Add files using upload-large-folder tool

real_inference/models/shared/torch/hub/checkpoints/resnet18-f37072fd.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f37072fd47e89c5e827621c5baffa7500819f7896bbacec160b1a16c560e07ec
+size 46830571

real_inference/source/XPolicyLab/policy/DP/diffusion_policy/common/checkpoint_util.py

@@ -0,0 +1,61 @@
+from typing import Optional, Dict
+import os
+
+
+class TopKCheckpointManager:
+
+    def __init__(
+        self,
+        save_dir,
+        monitor_key: str,
+        mode="min",
+        k=1,
+        format_str="epoch={epoch:03d}-train_loss={train_loss:.3f}.ckpt",
+    ):
+        assert mode in ["max", "min"]
+        assert k >= 0
+
+        self.save_dir = save_dir
+        self.monitor_key = monitor_key
+        self.mode = mode
+        self.k = k
+        self.format_str = format_str
+        self.path_value_map = dict()
+
+    def get_ckpt_path(self, data: Dict[str, float]) -> Optional[str]:
+        if self.k == 0:
+            return None
+
+        value = data[self.monitor_key]
+        ckpt_path = os.path.join(self.save_dir, self.format_str.format(**data))
+
+        if len(self.path_value_map) < self.k:
+            # under-capacity
+            self.path_value_map[ckpt_path] = value
+            return ckpt_path
+
+        # at capacity
+        sorted_map = sorted(self.path_value_map.items(), key=lambda x: x[1])
+        min_path, min_value = sorted_map[0]
+        max_path, max_value = sorted_map[-1]
+
+        delete_path = None
+        if self.mode == "max":
+            if value > min_value:
+                delete_path = min_path
+        else:
+            if value < max_value:
+                delete_path = max_path
+
+        if delete_path is None:
+            return None
+        else:
+            del self.path_value_map[delete_path]
+            self.path_value_map[ckpt_path] = value
+
+            if not os.path.exists(self.save_dir):
+                os.mkdir(self.save_dir)
+
+            if os.path.exists(delete_path):
+                os.remove(delete_path)
+            return ckpt_path

real_inference/source/XPolicyLab/policy/DP/diffusion_policy/common/cv2_util.py

@@ -0,0 +1,150 @@
+from typing import Tuple
+import math
+import cv2
+import numpy as np
+
+
+def draw_reticle(img, u, v, label_color):
+    """
+    Draws a reticle (cross-hair) on the image at the given position on top of
+    the original image.
+    @param img (In/Out) uint8 3 channel image
+    @param u X coordinate (width)
+    @param v Y coordinate (height)
+    @param label_color tuple of 3 ints for RGB color used for drawing.
+    """
+    # Cast to int.
+    u = int(u)
+    v = int(v)
+
+    white = (255, 255, 255)
+    cv2.circle(img, (u, v), 10, label_color, 1)
+    cv2.circle(img, (u, v), 11, white, 1)
+    cv2.circle(img, (u, v), 12, label_color, 1)
+    cv2.line(img, (u, v + 1), (u, v + 3), white, 1)
+    cv2.line(img, (u + 1, v), (u + 3, v), white, 1)
+    cv2.line(img, (u, v - 1), (u, v - 3), white, 1)
+    cv2.line(img, (u - 1, v), (u - 3, v), white, 1)
+
+
+def draw_text(
+        img,
+        *,
+        text,
+        uv_top_left,
+        color=(255, 255, 255),
+        fontScale=0.5,
+        thickness=1,
+        fontFace=cv2.FONT_HERSHEY_SIMPLEX,
+        outline_color=(0, 0, 0),
+        line_spacing=1.5,
+):
+    """
+    Draws multiline with an outline.
+    """
+    assert isinstance(text, str)
+
+    uv_top_left = np.array(uv_top_left, dtype=float)
+    assert uv_top_left.shape == (2, )
+
+    for line in text.splitlines():
+        (w, h), _ = cv2.getTextSize(
+            text=line,
+            fontFace=fontFace,
+            fontScale=fontScale,
+            thickness=thickness,
+        )
+        uv_bottom_left_i = uv_top_left + [0, h]
+        org = tuple(uv_bottom_left_i.astype(int))
+
+        if outline_color is not None:
+            cv2.putText(
+                img,
+                text=line,
+                org=org,
+                fontFace=fontFace,
+                fontScale=fontScale,
+                color=outline_color,
+                thickness=thickness * 3,
+                lineType=cv2.LINE_AA,
+            )
+        cv2.putText(
+            img,
+            text=line,
+            org=org,
+            fontFace=fontFace,
+            fontScale=fontScale,
+            color=color,
+            thickness=thickness,
+            lineType=cv2.LINE_AA,
+        )
+
+        uv_top_left += [0, h * line_spacing]
+
+
+def get_image_transform(
+        input_res: Tuple[int, int] = (1280, 720),
+        output_res: Tuple[int, int] = (640, 480),
+        bgr_to_rgb: bool = False,
+):
+
+    iw, ih = input_res
+    ow, oh = output_res
+    rw, rh = None, None
+    interp_method = cv2.INTER_AREA
+
+    if (iw / ih) >= (ow / oh):
+        # input is wider
+        rh = oh
+        rw = math.ceil(rh / ih * iw)
+        if oh > ih:
+            interp_method = cv2.INTER_LINEAR
+    else:
+        rw = ow
+        rh = math.ceil(rw / iw * ih)
+        if ow > iw:
+            interp_method = cv2.INTER_LINEAR
+
+    w_slice_start = (rw - ow) // 2
+    w_slice = slice(w_slice_start, w_slice_start + ow)
+    h_slice_start = (rh - oh) // 2
+    h_slice = slice(h_slice_start, h_slice_start + oh)
+    c_slice = slice(None)
+    if bgr_to_rgb:
+        c_slice = slice(None, None, -1)
+
+    def transform(img: np.ndarray):
+        assert img.shape == ((ih, iw, 3))
+        # resize
+        img = cv2.resize(img, (rw, rh), interpolation=interp_method)
+        # crop
+        img = img[h_slice, w_slice, c_slice]
+        return img
+
+    return transform
+
+
+def optimal_row_cols(n_cameras, in_wh_ratio, max_resolution=(1920, 1080)):
+    out_w, out_h = max_resolution
+    out_wh_ratio = out_w / out_h
+
+    n_rows = np.arange(n_cameras, dtype=np.int64) + 1
+    n_cols = np.ceil(n_cameras / n_rows).astype(np.int64)
+    cat_wh_ratio = in_wh_ratio * (n_cols / n_rows)
+    ratio_diff = np.abs(out_wh_ratio - cat_wh_ratio)
+    best_idx = np.argmin(ratio_diff)
+    best_n_row = n_rows[best_idx]
+    best_n_col = n_cols[best_idx]
+    best_cat_wh_ratio = cat_wh_ratio[best_idx]
+
+    rw, rh = None, None
+    if best_cat_wh_ratio >= out_wh_ratio:
+        # cat is wider
+        rw = math.floor(out_w / best_n_col)
+        rh = math.floor(rw / in_wh_ratio)
+    else:
+        rh = math.floor(out_h / best_n_row)
+        rw = math.floor(rh * in_wh_ratio)
+
+    # crop_resolution = (rw, rh)
+    return rw, rh, best_n_col, best_n_row

real_inference/source/XPolicyLab/policy/DP/diffusion_policy/common/env_util.py

@@ -0,0 +1,28 @@
+import cv2
+import numpy as np
+
+
+def render_env_video(env, states, actions=None):
+    observations = states
+    imgs = list()
+    for i in range(len(observations)):
+        state = observations[i]
+        env.set_state(state)
+        if i == 0:
+            env.set_state(state)
+        img = env.render()
+        # draw action
+        if actions is not None:
+            action = actions[i]
+            coord = (action / 512 * 96).astype(np.int32)
+            cv2.drawMarker(
+                img,
+                coord,
+                color=(255, 0, 0),
+                markerType=cv2.MARKER_CROSS,
+                markerSize=8,
+                thickness=1,
+            )
+        imgs.append(img)
+    imgs = np.array(imgs)
+    return imgs

real_inference/source/XPolicyLab/policy/DP/diffusion_policy/common/nested_dict_util.py

@@ -0,0 +1,34 @@
+import functools
+
+
+def nested_dict_map(f, x):
+    """
+    Map f over all leaf of nested dict x
+    """
+
+    if not isinstance(x, dict):
+        return f(x)
+    y = dict()
+    for key, value in x.items():
+        y[key] = nested_dict_map(f, value)
+    return y
+
+
+def nested_dict_reduce(f, x):
+    """
+    Map f over all values of nested dict x, and reduce to a single value
+    """
+    if not isinstance(x, dict):
+        return x
+
+    reduced_values = list()
+    for value in x.values():
+        reduced_values.append(nested_dict_reduce(f, value))
+    y = functools.reduce(f, reduced_values)
+    return y
+
+
+def nested_dict_check(f, x):
+    bool_dict = nested_dict_map(f, x)
+    result = nested_dict_reduce(lambda x, y: x and y, bool_dict)
+    return result

real_inference/source/XPolicyLab/policy/DP/diffusion_policy/common/normalize_util.py

@@ -0,0 +1,197 @@
+from diffusion_policy.model.common.normalizer import SingleFieldLinearNormalizer
+from diffusion_policy.common.pytorch_util import (
+    dict_apply,
+    dict_apply_reduce,
+    dict_apply_split,
+)
+import numpy as np
+
+
+def get_range_normalizer_from_stat(stat, output_max=1, output_min=-1, range_eps=1e-7):
+    # -1, 1 normalization
+    input_max = stat["max"]
+    input_min = stat["min"]
+    input_range = input_max - input_min
+    ignore_dim = input_range < range_eps
+    input_range[ignore_dim] = output_max - output_min
+    scale = (output_max - output_min) / input_range
+    offset = output_min - scale * input_min
+    offset[ignore_dim] = (output_max + output_min) / 2 - input_min[ignore_dim]
+
+    return SingleFieldLinearNormalizer.create_manual(scale=scale, offset=offset, input_stats_dict=stat)
+
+
+def get_image_range_normalizer():
+    scale = np.array([2], dtype=np.float32)
+    offset = np.array([-1], dtype=np.float32)
+    stat = {
+        "min": np.array([0], dtype=np.float32),
+        "max": np.array([1], dtype=np.float32),
+        "mean": np.array([0.5], dtype=np.float32),
+        "std": np.array([np.sqrt(1 / 12)], dtype=np.float32),
+    }
+    return SingleFieldLinearNormalizer.create_manual(scale=scale, offset=offset, input_stats_dict=stat)
+
+
+def get_identity_normalizer_from_stat(stat):
+    scale = np.ones_like(stat["min"])
+    offset = np.zeros_like(stat["min"])
+    return SingleFieldLinearNormalizer.create_manual(scale=scale, offset=offset, input_stats_dict=stat)
+
+
+def robomimic_abs_action_normalizer_from_stat(stat, rotation_transformer):
+    result = dict_apply_split(stat, lambda x: {"pos": x[..., :3], "rot": x[..., 3:6], "gripper": x[..., 6:]})
+
+    def get_pos_param_info(stat, output_max=1, output_min=-1, range_eps=1e-7):
+        # -1, 1 normalization
+        input_max = stat["max"]
+        input_min = stat["min"]
+        input_range = input_max - input_min
+        ignore_dim = input_range < range_eps
+        input_range[ignore_dim] = output_max - output_min
+        scale = (output_max - output_min) / input_range
+        offset = output_min - scale * input_min
+        offset[ignore_dim] = (output_max + output_min) / 2 - input_min[ignore_dim]
+
+        return {"scale": scale, "offset": offset}, stat
+
+    def get_rot_param_info(stat):
+        example = rotation_transformer.forward(stat["mean"])
+        scale = np.ones_like(example)
+        offset = np.zeros_like(example)
+        info = {
+            "max": np.ones_like(example),
+            "min": np.full_like(example, -1),
+            "mean": np.zeros_like(example),
+            "std": np.ones_like(example),
+        }
+        return {"scale": scale, "offset": offset}, info
+
+    def get_gripper_param_info(stat):
+        example = stat["max"]
+        scale = np.ones_like(example)
+        offset = np.zeros_like(example)
+        info = {
+            "max": np.ones_like(example),
+            "min": np.full_like(example, -1),
+            "mean": np.zeros_like(example),
+            "std": np.ones_like(example),
+        }
+        return {"scale": scale, "offset": offset}, info
+
+    pos_param, pos_info = get_pos_param_info(result["pos"])
+    rot_param, rot_info = get_rot_param_info(result["rot"])
+    gripper_param, gripper_info = get_gripper_param_info(result["gripper"])
+
+    param = dict_apply_reduce([pos_param, rot_param, gripper_param], lambda x: np.concatenate(x, axis=-1))
+    info = dict_apply_reduce([pos_info, rot_info, gripper_info], lambda x: np.concatenate(x, axis=-1))
+
+    return SingleFieldLinearNormalizer.create_manual(scale=param["scale"],
+                                                     offset=param["offset"],
+                                                     input_stats_dict=info)
+
+
+def robomimic_abs_action_only_normalizer_from_stat(stat):
+    result = dict_apply_split(stat, lambda x: {"pos": x[..., :3], "other": x[..., 3:]})
+
+    def get_pos_param_info(stat, output_max=1, output_min=-1, range_eps=1e-7):
+        # -1, 1 normalization
+        input_max = stat["max"]
+        input_min = stat["min"]
+        input_range = input_max - input_min
+        ignore_dim = input_range < range_eps
+        input_range[ignore_dim] = output_max - output_min
+        scale = (output_max - output_min) / input_range
+        offset = output_min - scale * input_min
+        offset[ignore_dim] = (output_max + output_min) / 2 - input_min[ignore_dim]
+
+        return {"scale": scale, "offset": offset}, stat
+
+    def get_other_param_info(stat):
+        example = stat["max"]
+        scale = np.ones_like(example)
+        offset = np.zeros_like(example)
+        info = {
+            "max": np.ones_like(example),
+            "min": np.full_like(example, -1),
+            "mean": np.zeros_like(example),
+            "std": np.ones_like(example),
+        }
+        return {"scale": scale, "offset": offset}, info
+
+    pos_param, pos_info = get_pos_param_info(result["pos"])
+    other_param, other_info = get_other_param_info(result["other"])
+
+    param = dict_apply_reduce([pos_param, other_param], lambda x: np.concatenate(x, axis=-1))
+    info = dict_apply_reduce([pos_info, other_info], lambda x: np.concatenate(x, axis=-1))
+
+    return SingleFieldLinearNormalizer.create_manual(scale=param["scale"],
+                                                     offset=param["offset"],
+                                                     input_stats_dict=info)
+
+
+def robomimic_abs_action_only_dual_arm_normalizer_from_stat(stat):
+    Da = stat["max"].shape[-1]
+    Dah = Da // 2
+    result = dict_apply_split(
+        stat,
+        lambda x: {
+            "pos0": x[..., :3],
+            "other0": x[..., 3:Dah],
+            "pos1": x[..., Dah:Dah + 3],
+            "other1": x[..., Dah + 3:],
+        },
+    )
+
+    def get_pos_param_info(stat, output_max=1, output_min=-1, range_eps=1e-7):
+        # -1, 1 normalization
+        input_max = stat["max"]
+        input_min = stat["min"]
+        input_range = input_max - input_min
+        ignore_dim = input_range < range_eps
+        input_range[ignore_dim] = output_max - output_min
+        scale = (output_max - output_min) / input_range
+        offset = output_min - scale * input_min
+        offset[ignore_dim] = (output_max + output_min) / 2 - input_min[ignore_dim]
+
+        return {"scale": scale, "offset": offset}, stat
+
+    def get_other_param_info(stat):
+        example = stat["max"]
+        scale = np.ones_like(example)
+        offset = np.zeros_like(example)
+        info = {
+            "max": np.ones_like(example),
+            "min": np.full_like(example, -1),
+            "mean": np.zeros_like(example),
+            "std": np.ones_like(example),
+        }
+        return {"scale": scale, "offset": offset}, info
+
+    pos0_param, pos0_info = get_pos_param_info(result["pos0"])
+    pos1_param, pos1_info = get_pos_param_info(result["pos1"])
+    other0_param, other0_info = get_other_param_info(result["other0"])
+    other1_param, other1_info = get_other_param_info(result["other1"])
+
+    param = dict_apply_reduce(
+        [pos0_param, other0_param, pos1_param, other1_param],
+        lambda x: np.concatenate(x, axis=-1),
+    )
+    info = dict_apply_reduce(
+        [pos0_info, other0_info, pos1_info, other1_info],
+        lambda x: np.concatenate(x, axis=-1),
+    )
+
+    return SingleFieldLinearNormalizer.create_manual(scale=param["scale"],
+                                                     offset=param["offset"],
+                                                     input_stats_dict=info)
+
+
+def array_to_stats(arr: np.ndarray):
+    stat = {
+        "min": np.min(arr, axis=0),
+        "max": np.max(arr, axis=0),
+        "mean": np.mean(arr, axis=0),
+        "std": np.std(arr, axis=0),
+    }
+    return stat

real_inference/source/XPolicyLab/policy/DP/diffusion_policy/common/replay_buffer.py

@@ -0,0 +1,622 @@
+from typing import Union, Dict, Optional
+import os
+import math
+import numbers
+import zarr
+import numcodecs
+import numpy as np
+from functools import cached_property
+
+
+def check_chunks_compatible(chunks: tuple, shape: tuple):
+    assert len(shape) == len(chunks)
+    for c in chunks:
+        assert isinstance(c, numbers.Integral)
+        assert c > 0
+
+
+def rechunk_recompress_array(group, name, chunks=None, chunk_length=None, compressor=None, tmp_key="_temp"):
+    old_arr = group[name]
+    if chunks is None:
+        if chunk_length is not None:
+            chunks = (chunk_length, ) + old_arr.chunks[1:]
+        else:
+            chunks = old_arr.chunks
+    check_chunks_compatible(chunks, old_arr.shape)
+
+    if compressor is None:
+        compressor = old_arr.compressor
+
+    if (chunks == old_arr.chunks) and (compressor == old_arr.compressor):
+        # no change
+        return old_arr
+
+    # rechunk recompress
+    group.move(name, tmp_key)
+    old_arr = group[tmp_key]
+    n_copied, n_skipped, n_bytes_copied = zarr.copy(
+        source=old_arr,
+        dest=group,
+        name=name,
+        chunks=chunks,
+        compressor=compressor,
+    )
+    del group[tmp_key]
+    arr = group[name]
+    return arr
+
+
+def get_optimal_chunks(shape, dtype, target_chunk_bytes=2e6, max_chunk_length=None):
+    """
+    Common shapes
+    T,D
+    T,N,D
+    T,H,W,C
+    T,N,H,W,C
+    """
+    itemsize = np.dtype(dtype).itemsize
+    # reversed
+    rshape = list(shape[::-1])
+    if max_chunk_length is not None:
+        rshape[-1] = int(max_chunk_length)
+    split_idx = len(shape) - 1
+    for i in range(len(shape) - 1):
+        this_chunk_bytes = itemsize * np.prod(rshape[:i])
+        next_chunk_bytes = itemsize * np.prod(rshape[:i + 1])
+        if (this_chunk_bytes <= target_chunk_bytes and next_chunk_bytes > target_chunk_bytes):
+            split_idx = i
+
+    rchunks = rshape[:split_idx]
+    item_chunk_bytes = itemsize * np.prod(rshape[:split_idx])
+    this_max_chunk_length = rshape[split_idx]
+    next_chunk_length = min(this_max_chunk_length, math.ceil(target_chunk_bytes / item_chunk_bytes))
+    rchunks.append(next_chunk_length)
+    len_diff = len(shape) - len(rchunks)
+    rchunks.extend([1] * len_diff)
+    chunks = tuple(rchunks[::-1])
+    # print(np.prod(chunks) * itemsize / target_chunk_bytes)
+    return chunks
+
+
+class ReplayBuffer:
+    """
+    Zarr-based temporal datastructure.
+    Assumes first dimension to be time. Only chunk in time dimension.
+    """
+
+    def __init__(self, root: Union[zarr.Group, Dict[str, dict]]):
+        """
+        Dummy constructor. Use copy_from* and create_from* class methods instead.
+        """
+        assert "data" in root
+        assert "meta" in root
+        assert "episode_ends" in root["meta"]
+        for key, value in root["data"].items():
+            assert value.shape[0] == root["meta"]["episode_ends"][-1]
+        self.root = root
+
+    # ============= create constructors ===============
+    @classmethod
+    def create_empty_zarr(cls, storage=None, root=None):
+        if root is None:
+            if storage is None:
+                storage = zarr.MemoryStore()
+            root = zarr.group(store=storage)
+        data = root.require_group("data", overwrite=False)
+        meta = root.require_group("meta", overwrite=False)
+        if "episode_ends" not in meta:
+            episode_ends = meta.zeros(
+                "episode_ends",
+                shape=(0, ),
+                dtype=np.int64,
+                compressor=None,
+                overwrite=False,
+            )
+        return cls(root=root)
+
+    @classmethod
+    def create_empty_numpy(cls):
+        root = {
+            "data": dict(),
+            "meta": {
+                "episode_ends": np.zeros((0, ), dtype=np.int64)
+            },
+        }
+        return cls(root=root)
+
+    @classmethod
+    def create_from_group(cls, group, **kwargs):
+        if "data" not in group:
+            # create from stratch
+            buffer = cls.create_empty_zarr(root=group, **kwargs)
+        else:
+            # already exist
+            buffer = cls(root=group, **kwargs)
+        return buffer
+
+    @classmethod
+    def create_from_path(cls, zarr_path, mode="r", **kwargs):
+        """
+        Open a on-disk zarr directly (for dataset larger than memory).
+        Slower.
+        """
+        group = zarr.open(os.path.expanduser(zarr_path), mode)
+        return cls.create_from_group(group, **kwargs)
+
+    # ============= copy constructors ===============
+    @classmethod
+    def copy_from_store(
+            cls,
+            src_store,
+            store=None,
+            keys=None,
+            chunks: Dict[str, tuple] = dict(),
+            compressors: Union[dict, str, numcodecs.abc.Codec] = dict(),
+            if_exists="replace",
+            **kwargs,
+    ):
+        """
+        Load to memory.
+        """
+        src_root = zarr.group(src_store)
+        root = None
+        if store is None:
+            # numpy backend
+            meta = dict()
+            for key, value in src_root["meta"].items():
+                if len(value.shape) == 0:
+                    meta[key] = np.array(value)
+                else:
+                    meta[key] = value[:]
+
+            if keys is None:
+                keys = src_root["data"].keys()
+            data = dict()
+            for key in keys:
+                arr = src_root["data"][key]
+                data[key] = arr[:]
+
+            root = {"meta": meta, "data": data}
+        else:
+            root = zarr.group(store=store)
+            # copy without recompression
+            n_copied, n_skipped, n_bytes_copied = zarr.copy_store(
+                source=src_store,
+                dest=store,
+                source_path="/meta",
+                dest_path="/meta",
+                if_exists=if_exists,
+            )
+            data_group = root.create_group("data", overwrite=True)
+            if keys is None:
+                keys = src_root["data"].keys()
+            for key in keys:
+                value = src_root["data"][key]
+                cks = cls._resolve_array_chunks(chunks=chunks, key=key, array=value)
+                cpr = cls._resolve_array_compressor(compressors=compressors, key=key, array=value)
+                if cks == value.chunks and cpr == value.compressor:
+                    # copy without recompression
+                    this_path = "/data/" + key
+                    n_copied, n_skipped, n_bytes_copied = zarr.copy_store(
+                        source=src_store,
+                        dest=store,
+                        source_path=this_path,
+                        dest_path=this_path,
+                        if_exists=if_exists,
+                    )
+                else:
+                    # copy with recompression
+                    n_copied, n_skipped, n_bytes_copied = zarr.copy(
+                        source=value,
+                        dest=data_group,
+                        name=key,
+                        chunks=cks,
+                        compressor=cpr,
+                        if_exists=if_exists,
+                    )
+        buffer = cls(root=root)
+        return buffer
+
+    @classmethod
+    def copy_from_path(
+            cls,
+            zarr_path,
+            backend=None,
+            store=None,
+            keys=None,
+            chunks: Dict[str, tuple] = dict(),
+            compressors: Union[dict, str, numcodecs.abc.Codec] = dict(),
+            if_exists="replace",
+            **kwargs,
+    ):
+        """
+        Copy a on-disk zarr to in-memory compressed.
+        Recommended
+        """
+        if backend == "numpy":
+            print("backend argument is deprecated!")
+            store = None
+        group = zarr.open(os.path.expanduser(zarr_path), "r")
+        return cls.copy_from_store(
+            src_store=group.store,
+            store=store,
+            keys=keys,
+            chunks=chunks,
+            compressors=compressors,
+            if_exists=if_exists,
+            **kwargs,
+        )
+
+    # ============= save methods ===============
+    def save_to_store(
+            self,
+            store,
+            chunks: Optional[Dict[str, tuple]] = dict(),
+            compressors: Union[str, numcodecs.abc.Codec, dict] = dict(),
+            if_exists="replace",
+            **kwargs,
+    ):
+
+        root = zarr.group(store)
+        if self.backend == "zarr":
+            # recompression free copy
+            n_copied, n_skipped, n_bytes_copied = zarr.copy_store(
+                source=self.root.store,
+                dest=store,
+                source_path="/meta",
+                dest_path="/meta",
+                if_exists=if_exists,
+            )
+        else:
+            meta_group = root.create_group("meta", overwrite=True)
+            # save meta, no chunking
+            for key, value in self.root["meta"].items():
+                _ = meta_group.array(name=key, data=value, shape=value.shape, chunks=value.shape)
+
+        # save data, chunk
+        data_group = root.create_group("data", overwrite=True)
+        for key, value in self.root["data"].items():
+            cks = self._resolve_array_chunks(chunks=chunks, key=key, array=value)
+            cpr = self._resolve_array_compressor(compressors=compressors, key=key, array=value)
+            if isinstance(value, zarr.Array):
+                if cks == value.chunks and cpr == value.compressor:
+                    # copy without recompression
+                    this_path = "/data/" + key
+                    n_copied, n_skipped, n_bytes_copied = zarr.copy_store(
+                        source=self.root.store,
+                        dest=store,
+                        source_path=this_path,
+                        dest_path=this_path,
+                        if_exists=if_exists,
+                    )
+                else:
+                    # copy with recompression
+                    n_copied, n_skipped, n_bytes_copied = zarr.copy(
+                        source=value,
+                        dest=data_group,
+                        name=key,
+                        chunks=cks,
+                        compressor=cpr,
+                        if_exists=if_exists,
+                    )
+            else:
+                # numpy
+                _ = data_group.array(name=key, data=value, chunks=cks, compressor=cpr)
+        return store
+
+    def save_to_path(
+            self,
+            zarr_path,
+            chunks: Optional[Dict[str, tuple]] = dict(),
+            compressors: Union[str, numcodecs.abc.Codec, dict] = dict(),
+            if_exists="replace",
+            **kwargs,
+    ):
+        store = zarr.DirectoryStore(os.path.expanduser(zarr_path))
+        return self.save_to_store(store, chunks=chunks, compressors=compressors, if_exists=if_exists, **kwargs)
+
+    @staticmethod
+    def resolve_compressor(compressor="default"):
+        if compressor == "default":
+            compressor = numcodecs.Blosc(cname="lz4", clevel=5, shuffle=numcodecs.Blosc.NOSHUFFLE)
+        elif compressor == "disk":
+            compressor = numcodecs.Blosc("zstd", clevel=5, shuffle=numcodecs.Blosc.BITSHUFFLE)
+        return compressor
+
+    @classmethod
+    def _resolve_array_compressor(cls, compressors: Union[dict, str, numcodecs.abc.Codec], key, array):
+        # allows compressor to be explicitly set to None
+        cpr = "nil"
+        if isinstance(compressors, dict):
+            if key in compressors:
+                cpr = cls.resolve_compressor(compressors[key])
+            elif isinstance(array, zarr.Array):
+                cpr = array.compressor
+        else:
+            cpr = cls.resolve_compressor(compressors)
+        # backup default
+        if cpr == "nil":
+            cpr = cls.resolve_compressor("default")
+        return cpr
+
+    @classmethod
+    def _resolve_array_chunks(cls, chunks: Union[dict, tuple], key, array):
+        cks = None
+        if isinstance(chunks, dict):
+            if key in chunks:
+                cks = chunks[key]
+            elif isinstance(array, zarr.Array):
+                cks = array.chunks
+        elif isinstance(chunks, tuple):
+            cks = chunks
+        else:
+            raise TypeError(f"Unsupported chunks type {type(chunks)}")
+        # backup default
+        if cks is None:
+            cks = get_optimal_chunks(shape=array.shape, dtype=array.dtype)
+        # check
+        check_chunks_compatible(chunks=cks, shape=array.shape)
+        return cks
+
+    # ============= properties =================
+    @cached_property
+    def data(self):
+        return self.root["data"]
+
+    @cached_property
+    def meta(self):
+        return self.root["meta"]
+
+    def update_meta(self, data):
+        # sanitize data
+        np_data = dict()
+        for key, value in data.items():
+            if isinstance(value, np.ndarray):
+                np_data[key] = value
+            else:
+                arr = np.array(value)
+                if arr.dtype == object:
+                    raise TypeError(f"Invalid value type {type(value)}")
+                np_data[key] = arr
+
+        meta_group = self.meta
+        if self.backend == "zarr":
+            for key, value in np_data.items():
+                _ = meta_group.array(
+                    name=key,
+                    data=value,
+                    shape=value.shape,
+                    chunks=value.shape,
+                    overwrite=True,
+                )
+        else:
+            meta_group.update(np_data)
+
+        return meta_group
+
+    @property
+    def episode_ends(self):
+        return self.meta["episode_ends"]
+
+    def get_episode_idxs(self):
+        import numba
+
+        numba.jit(nopython=True)
+
+        def _get_episode_idxs(episode_ends):
+            result = np.zeros((episode_ends[-1], ), dtype=np.int64)
+            for i in range(len(episode_ends)):
+                start = 0
+                if i > 0:
+                    start = episode_ends[i - 1]
+                end = episode_ends[i]
+                for idx in range(start, end):
+                    result[idx] = i
+            return result
+
+        return _get_episode_idxs(self.episode_ends)
+
+    @property
+    def backend(self):
+        backend = "numpy"
+        if isinstance(self.root, zarr.Group):
+            backend = "zarr"
+        return backend
+
+    # =========== dict-like API ==============
+    def __repr__(self) -> str:
+        if self.backend == "zarr":
+            return str(self.root.tree())
+        else:
+            return super().__repr__()
+
+    def keys(self):
+        return self.data.keys()
+
+    def values(self):
+        return self.data.values()
+
+    def items(self):
+        return self.data.items()
+
+    def __getitem__(self, key):
+        return self.data[key]
+
+    def __contains__(self, key):
+        return key in self.data
+
+    # =========== our API ==============
+    @property
+    def n_steps(self):
+        if len(self.episode_ends) == 0:
+            return 0
+        return self.episode_ends[-1]
+
+    @property
+    def n_episodes(self):
+        return len(self.episode_ends)
+
+    @property
+    def chunk_size(self):
+        if self.backend == "zarr":
+            return next(iter(self.data.arrays()))[-1].chunks[0]
+        return None
+
+    @property
+    def episode_lengths(self):
+        ends = self.episode_ends[:]
+        ends = np.insert(ends, 0, 0)
+        lengths = np.diff(ends)
+        return lengths
+
+    def add_episode(
+            self,
+            data: Dict[str, np.ndarray],
+            chunks: Optional[Dict[str, tuple]] = dict(),
+            compressors: Union[str, numcodecs.abc.Codec, dict] = dict(),
+    ):
+        assert len(data) > 0
+        is_zarr = self.backend == "zarr"
+
+        curr_len = self.n_steps
+        episode_length = None
+        for key, value in data.items():
+            assert len(value.shape) >= 1
+            if episode_length is None:
+                episode_length = len(value)
+            else:
+                assert episode_length == len(value)
+        new_len = curr_len + episode_length
+
+        for key, value in data.items():
+            new_shape = (new_len, ) + value.shape[1:]
+            # create array
+            if key not in self.data:
+                if is_zarr:
+                    cks = self._resolve_array_chunks(chunks=chunks, key=key, array=value)
+                    cpr = self._resolve_array_compressor(compressors=compressors, key=key, array=value)
+                    arr = self.data.zeros(
+                        name=key,
+                        shape=new_shape,
+                        chunks=cks,
+                        dtype=value.dtype,
+                        compressor=cpr,
+                    )
+                else:
+                    # copy data to prevent modify
+                    arr = np.zeros(shape=new_shape, dtype=value.dtype)
+                    self.data[key] = arr
+            else:
+                arr = self.data[key]
+                assert value.shape[1:] == arr.shape[1:]
+                # same method for both zarr and numpy
+                if is_zarr:
+                    arr.resize(new_shape)
+                else:
+                    arr.resize(new_shape, refcheck=False)
+            # copy data
+            arr[-value.shape[0]:] = value
+
+        # append to episode ends
+        episode_ends = self.episode_ends
+        if is_zarr:
+            episode_ends.resize(episode_ends.shape[0] + 1)
+        else:
+            episode_ends.resize(episode_ends.shape[0] + 1, refcheck=False)
+        episode_ends[-1] = new_len
+
+        # rechunk
+        if is_zarr:
+            if episode_ends.chunks[0] < episode_ends.shape[0]:
+                rechunk_recompress_array(
+                    self.meta,
+                    "episode_ends",
+                    chunk_length=int(episode_ends.shape[0] * 1.5),
+                )
+
+    def drop_episode(self):
+        is_zarr = self.backend == "zarr"
+        episode_ends = self.episode_ends[:].copy()
+        assert len(episode_ends) > 0
+        start_idx = 0
+        if len(episode_ends) > 1:
+            start_idx = episode_ends[-2]
+        for key, value in self.data.items():
+            new_shape = (start_idx, ) + value.shape[1:]
+            if is_zarr:
+                value.resize(new_shape)
+            else:
+                value.resize(new_shape, refcheck=False)
+        if is_zarr:
+            self.episode_ends.resize(len(episode_ends) - 1)
+        else:
+            self.episode_ends.resize(len(episode_ends) - 1, refcheck=False)
+
+    def pop_episode(self):
+        assert self.n_episodes > 0
+        episode = self.get_episode(self.n_episodes - 1, copy=True)
+        self.drop_episode()
+        return episode
+
+    def extend(self, data):
+        self.add_episode(data)
+
+    def get_episode(self, idx, copy=False):
+        idx = list(range(len(self.episode_ends)))[idx]
+        start_idx = 0
+        if idx > 0:
+            start_idx = self.episode_ends[idx - 1]
+        end_idx = self.episode_ends[idx]
+        result = self.get_steps_slice(start_idx, end_idx, copy=copy)
+        return result
+
+    def get_episode_slice(self, idx):
+        start_idx = 0
+        if idx > 0:
+            start_idx = self.episode_ends[idx - 1]
+        end_idx = self.episode_ends[idx]
+        return slice(start_idx, end_idx)
+
+    def get_steps_slice(self, start, stop, step=None, copy=False):
+        _slice = slice(start, stop, step)
+
+        result = dict()
+        for key, value in self.data.items():
+            x = value[_slice]
+            if copy and isinstance(value, np.ndarray):
+                x = x.copy()
+            result[key] = x
+        return result
+
+    # =========== chunking =============
+    def get_chunks(self) -> dict:
+        assert self.backend == "zarr"
+        chunks = dict()
+        for key, value in self.data.items():
+            chunks[key] = value.chunks
+        return chunks
+
+    def set_chunks(self, chunks: dict):
+        assert self.backend == "zarr"
+        for key, value in chunks.items():
+            if key in self.data:
+                arr = self.data[key]
+                if value != arr.chunks:
+                    check_chunks_compatible(chunks=value, shape=arr.shape)
+                    rechunk_recompress_array(self.data, key, chunks=value)
+
+    def get_compressors(self) -> dict:
+        assert self.backend == "zarr"
+        compressors = dict()
+        for key, value in self.data.items():
+            compressors[key] = value.compressor
+        return compressors
+
+    def set_compressors(self, compressors: dict):
+        assert self.backend == "zarr"
+        for key, value in compressors.items():
+            if key in self.data:
+                arr = self.data[key]
+                compressor = self.resolve_compressor(value)
+                if compressor != arr.compressor:
+                    rechunk_recompress_array(self.data, key, compressor=compressor)

real_inference/source/XPolicyLab/policy/DP/diffusion_policy/common/robomimic_config_util.py

@@ -0,0 +1,41 @@
+from omegaconf import OmegaConf
+from robomimic.config import config_factory
+import robomimic.scripts.generate_paper_configs as gpc
+from robomimic.scripts.generate_paper_configs import (
+    modify_config_for_default_image_exp,
+    modify_config_for_default_low_dim_exp,
+    modify_config_for_dataset,
+)
+
+
+def get_robomimic_config(algo_name="bc_rnn", hdf5_type="low_dim", task_name="square", dataset_type="ph"):
+    base_dataset_dir = "/tmp/null"
+    filter_key = None
+
+    # decide whether to use low-dim or image training defaults
+    modifier_for_obs = modify_config_for_default_image_exp
+    if hdf5_type in ["low_dim", "low_dim_sparse", "low_dim_dense"]:
+        modifier_for_obs = modify_config_for_default_low_dim_exp
+
+    algo_config_name = "bc" if algo_name == "bc_rnn" else algo_name
+    config = config_factory(algo_name=algo_config_name)
+    # turn into default config for observation modalities (e.g.: low-dim or rgb)
+    config = modifier_for_obs(config)
+    # add in config based on the dataset
+    config = modify_config_for_dataset(
+        config=config,
+        task_name=task_name,
+        dataset_type=dataset_type,
+        hdf5_type=hdf5_type,
+        base_dataset_dir=base_dataset_dir,
+        filter_key=filter_key,
+    )
+    # add in algo hypers based on dataset
+    algo_config_modifier = getattr(gpc, f"modify_{algo_name}_config_for_dataset")
+    config = algo_config_modifier(
+        config=config,
+        task_name=task_name,
+        dataset_type=dataset_type,
+        hdf5_type=hdf5_type,
+    )
+    return config

real_inference/source/XPolicyLab/policy/DP/diffusion_policy/common/timestamp_accumulator.py

@@ -0,0 +1,220 @@
+from typing import List, Tuple, Optional, Dict
+import math
+import numpy as np
+
+
+def get_accumulate_timestamp_idxs(
+    timestamps: List[float],
+    start_time: float,
+    dt: float,
+    eps: float = 1e-5,
+    next_global_idx: Optional[int] = 0,
+    allow_negative=False,
+) -> Tuple[List[int], List[int], int]:
+    """
+    For each dt window, choose the first timestamp in the window.
+    Assumes timestamps sorted. One timestamp might be chosen multiple times due to dropped frames.
+    next_global_idx should start at 0 normally, and then use the returned next_global_idx.
+    However, when overwiting previous values are desired, set last_global_idx to None.
+
+    Returns:
+    local_idxs: which index in the given timestamps array to chose from
+    global_idxs: the global index of each chosen timestamp
+    next_global_idx: used for next call.
+    """
+    local_idxs = list()
+    global_idxs = list()
+    for local_idx, ts in enumerate(timestamps):
+        # add eps * dt to timestamps so that when ts == start_time + k * dt
+        # is always recorded as kth element (avoiding floating point errors)
+        global_idx = math.floor((ts - start_time) / dt + eps)
+        if (not allow_negative) and (global_idx < 0):
+            continue
+        if next_global_idx is None:
+            next_global_idx = global_idx
+
+        n_repeats = max(0, global_idx - next_global_idx + 1)
+        for i in range(n_repeats):
+            local_idxs.append(local_idx)
+            global_idxs.append(next_global_idx + i)
+        next_global_idx += n_repeats
+    return local_idxs, global_idxs, next_global_idx
+
+
+def align_timestamps(
+    timestamps: List[float],
+    target_global_idxs: List[int],
+    start_time: float,
+    dt: float,
+    eps: float = 1e-5,
+):
+    if isinstance(target_global_idxs, np.ndarray):
+        target_global_idxs = target_global_idxs.tolist()
+    assert len(target_global_idxs) > 0
+
+    local_idxs, global_idxs, _ = get_accumulate_timestamp_idxs(
+        timestamps=timestamps,
+        start_time=start_time,
+        dt=dt,
+        eps=eps,
+        next_global_idx=target_global_idxs[0],
+        allow_negative=True,
+    )
+    if len(global_idxs) > len(target_global_idxs):
+        # if more steps available, truncate
+        global_idxs = global_idxs[:len(target_global_idxs)]
+        local_idxs = local_idxs[:len(target_global_idxs)]
+
+    if len(global_idxs) == 0:
+        import pdb
+
+        pdb.set_trace()
+
+    for i in range(len(target_global_idxs) - len(global_idxs)):
+        # if missing, repeat
+        local_idxs.append(len(timestamps) - 1)
+        global_idxs.append(global_idxs[-1] + 1)
+    assert global_idxs == target_global_idxs
+    assert len(local_idxs) == len(global_idxs)
+    return local_idxs
+
+
+class TimestampObsAccumulator:
+
+    def __init__(self, start_time: float, dt: float, eps: float = 1e-5):
+        self.start_time = start_time
+        self.dt = dt
+        self.eps = eps
+        self.obs_buffer = dict()
+        self.timestamp_buffer = None
+        self.next_global_idx = 0
+
+    def __len__(self):
+        return self.next_global_idx
+
+    @property
+    def data(self):
+        if self.timestamp_buffer is None:
+            return dict()
+        result = dict()
+        for key, value in self.obs_buffer.items():
+            result[key] = value[:len(self)]
+        return result
+
+    @property
+    def actual_timestamps(self):
+        if self.timestamp_buffer is None:
+            return np.array([])
+        return self.timestamp_buffer[:len(self)]
+
+    @property
+    def timestamps(self):
+        if self.timestamp_buffer is None:
+            return np.array([])
+        return self.start_time + np.arange(len(self)) * self.dt
+
+    def put(self, data: Dict[str, np.ndarray], timestamps: np.ndarray):
+        """
+        data:
+            key: T,*
+        """
+
+        local_idxs, global_idxs, self.next_global_idx = get_accumulate_timestamp_idxs(
+            timestamps=timestamps,
+            start_time=self.start_time,
+            dt=self.dt,
+            eps=self.eps,
+            next_global_idx=self.next_global_idx,
+        )
+
+        if len(global_idxs) > 0:
+            if self.timestamp_buffer is None:
+                # first allocation
+                self.obs_buffer = dict()
+                for key, value in data.items():
+                    self.obs_buffer[key] = np.zeros_like(value)
+                self.timestamp_buffer = np.zeros((len(timestamps), ), dtype=np.float64)
+
+            this_max_size = global_idxs[-1] + 1
+            if this_max_size > len(self.timestamp_buffer):
+                # reallocate
+                new_size = max(this_max_size, len(self.timestamp_buffer) * 2)
+                for key in list(self.obs_buffer.keys()):
+                    new_shape = (new_size, ) + self.obs_buffer[key].shape[1:]
+                    self.obs_buffer[key] = np.resize(self.obs_buffer[key], new_shape)
+                self.timestamp_buffer = np.resize(self.timestamp_buffer, (new_size))
+
+            # write data
+            for key, value in self.obs_buffer.items():
+                value[global_idxs] = data[key][local_idxs]
+            self.timestamp_buffer[global_idxs] = timestamps[local_idxs]
+
+
+class TimestampActionAccumulator:
+
+    def __init__(self, start_time: float, dt: float, eps: float = 1e-5):
+        """
+        Different from Obs accumulator, the action accumulator
+        allows overwriting previous values.
+        """
+        self.start_time = start_time
+        self.dt = dt
+        self.eps = eps
+        self.action_buffer = None
+        self.timestamp_buffer = None
+        self.size = 0
+
+    def __len__(self):
+        return self.size
+
+    @property
+    def actions(self):
+        if self.action_buffer is None:
+            return np.array([])
+        return self.action_buffer[:len(self)]
+
+    @property
+    def actual_timestamps(self):
+        if self.timestamp_buffer is None:
+            return np.array([])
+        return self.timestamp_buffer[:len(self)]
+
+    @property
+    def timestamps(self):
+        if self.timestamp_buffer is None:
+            return np.array([])
+        return self.start_time + np.arange(len(self)) * self.dt
+
+    def put(self, actions: np.ndarray, timestamps: np.ndarray):
+        """
+        Note: timestamps is the time when the action will be issued,
+        not when the action will be completed (target_timestamp)
+        """
+
+        local_idxs, global_idxs, _ = get_accumulate_timestamp_idxs(
+            timestamps=timestamps,
+            start_time=self.start_time,
+            dt=self.dt,
+            eps=self.eps,
+            # allows overwriting previous actions
+            next_global_idx=None,
+        )
+
+        if len(global_idxs) > 0:
+            if self.timestamp_buffer is None:
+                # first allocation
+                self.action_buffer = np.zeros_like(actions)
+                self.timestamp_buffer = np.zeros((len(actions), ), dtype=np.float64)
+
+            this_max_size = global_idxs[-1] + 1
+            if this_max_size > len(self.timestamp_buffer):
+                # reallocate
+                new_size = max(this_max_size, len(self.timestamp_buffer) * 2)
+                new_shape = (new_size, ) + self.action_buffer.shape[1:]
+                self.action_buffer = np.resize(self.action_buffer, new_shape)
+                self.timestamp_buffer = np.resize(self.timestamp_buffer, (new_size, ))
+
+            # potentially rewrite old data (as expected)
+            self.action_buffer[global_idxs] = actions[local_idxs]
+            self.timestamp_buffer[global_idxs] = timestamps[local_idxs]
+            self.size = max(self.size, this_max_size)

real_inference/source/XPolicyLab/policy/DP/diffusion_policy/env_runner/dp_runner.py

@@ -0,0 +1,101 @@
+import torch
+import numpy as np
+from collections import deque
+
+from diffusion_policy.common.pytorch_util import dict_apply
+from diffusion_policy.policy.base_image_policy import BaseImagePolicy
+
+class DPRunner:
+
+    def __init__(
+        self,
+        eval_episodes=20,
+        max_steps=300,
+        n_obs_steps=3,
+        n_action_steps=8,
+        fps=10,
+        crf=22,
+        tqdm_interval_sec=5.0,
+        task_name=None,
+    ):
+        self.task_name = task_name
+        self.eval_episodes = eval_episodes
+        self.fps = fps
+        self.crf = crf
+        self.n_obs_steps = n_obs_steps
+        self.n_action_steps = n_action_steps
+        self.max_steps = max_steps
+        self.tqdm_interval_sec = tqdm_interval_sec
+
+        self.obs_list = [deque(maxlen=n_obs_steps + 1) for _ in range(100)] # TODO
+
+    def stack_last_n_obs(self, all_obs, n_steps):
+        assert len(all_obs) > 0
+        all_obs = list(all_obs)
+        if isinstance(all_obs[0], np.ndarray):
+            result = np.zeros((n_steps, ) + all_obs[-1].shape, dtype=all_obs[-1].dtype)
+            start_idx = -min(n_steps, len(all_obs))
+            result[start_idx:] = np.array(all_obs[start_idx:])
+            if n_steps > len(all_obs):
+                # pad
+                result[:start_idx] = result[start_idx]
+        elif isinstance(all_obs[0], torch.Tensor):
+            result = torch.zeros((n_steps, ) + all_obs[-1].shape, dtype=all_obs[-1].dtype)
+            start_idx = -min(n_steps, len(all_obs))
+            result[start_idx:] = torch.stack(all_obs[start_idx:])
+            if n_steps > len(all_obs):
+                # pad
+                result[:start_idx] = result[start_idx]
+        else:
+            raise RuntimeError(f"Unsupported obs type {type(all_obs[0])}")
+        return result
+
+    def reset_obs(self):
+        for q in self.obs_list:
+            q.clear()
+
+    def update_obs(self, current_obs_list, env_idx_list):
+        for env_idx, current_obs in zip(env_idx_list, current_obs_list):
+            self.obs_list[env_idx].append(current_obs)
+
+    def get_n_steps_obs(self, env_idx_list):
+        result_list = []
+        for env_idx in env_idx_list:
+            result = dict()
+            for key in self.obs_list[env_idx][0].keys():
+                result[key] = self.stack_last_n_obs([obs[key] for obs in self.obs_list[env_idx]], self.n_obs_steps)
+            result_list.append(result)
+        return result_list
+
+    def get_action(self, policy: BaseImagePolicy, env_idx_list):
+        device = policy.device
+        obs_list = self.get_n_steps_obs(env_idx_list)
+
+        # device transfer
+
+        obs_list_numpy = [
+            dict_apply(obs, lambda x: torch.from_numpy(x).to(device=device))
+            for obs in obs_list
+        ]
+
+        with torch.no_grad():
+            obs_dict_input = {}
+            obs_dict_input["head_cam"] = torch.stack(
+                [obs_torch["head_cam"] for obs_torch in obs_list_numpy], dim=0
+            )
+            obs_dict_input["left_cam"] = torch.stack(
+                [obs_torch["left_cam"] for obs_torch in obs_list_numpy], dim=0
+            )
+            obs_dict_input["right_cam"] = torch.stack(
+                [obs_torch["right_cam"] for obs_torch in obs_list_numpy], dim=0
+            )
+            obs_dict_input["agent_pos"] = torch.stack(
+                [obs_torch["agent_pos"] for obs_torch in obs_list_numpy], dim=0
+            )
+
+            action_dict = policy.predict_action(obs_dict_input)
+
+        np_action_dict = dict_apply(action_dict, lambda x: x.detach().cpu().numpy())
+        actions = np_action_dict["action"][:, :self.n_action_steps]
+
+        return actions

real_inference/source/XPolicyLab/policy/DP/diffusion_policy/process_data.py

@@ -0,0 +1,122 @@
+import os
+import glob
+import numpy as np
+import zarr
+import shutil
+import argparse
+import cv2
+from XPolicyLab.utils.load_file import load_hdf5
+from XPolicyLab.utils.process_data import pack_robot_state, get_robot_action_dim_info, decode_image_bit
+
+def main():
+    parser = argparse.ArgumentParser(description="Process some episodes.")
+    parser.add_argument("bench_name", type=str, help="The name of the benchmark (e.g., RoboDojo)",)
+    parser.add_argument("ckpt_name", type=str, help="Run name; also selects raw task dir under data/<bench>/",)
+    parser.add_argument("env_cfg_type", type=str, help="The name of the environment config",)
+    parser.add_argument("action_type", type=str, help="The type of action to process (e.g., joint)",)
+    parser.add_argument("expert_data_num", type=int, nargs="?", default=None,
+                        help="Optional number of episodes to process; defaults to all episodes.",)
+    args = parser.parse_args()
+
+    bench_name = args.bench_name
+    ckpt_name = args.ckpt_name
+    env_cfg_type = args.env_cfg_type
+    expert_data_num = args.expert_data_num
+    action_type = args.action_type
+    load_data_dir = os.path.join("../../../data", str(bench_name), str(ckpt_name), str(env_cfg_type))
+
+    robot_action_dim_info = get_robot_action_dim_info(env_cfg_type)
+
+    available_episodes = len(glob.glob(os.path.join(load_data_dir, "data", "episode_*.hdf5")))
+    if expert_data_num is None:
+        expert_data_num = available_episodes
+    assert expert_data_num <= available_episodes, (
+        f"Requested {expert_data_num} episodes but only {available_episodes} available in {load_data_dir}")
+
+    frame_count = 0
+
+    save_dir = f"./data/{bench_name}-{ckpt_name}-{env_cfg_type}-{action_type}.zarr"
+
+    if os.path.exists(save_dir):
+        shutil.rmtree(save_dir)
+
+    current_episode = 0
+
+    zarr_root = zarr.group(save_dir)
+    zarr_data = zarr_root.create_group("data")
+    zarr_meta = zarr_root.create_group("meta")
+
+    head_camera_arrays, left_camera_arrays, right_camera_arrays = [], [], []
+    episode_ends_arrays, action_arrays, state_arrays = ([], [], [],)
+
+    while current_episode < expert_data_num:
+        print(f"DP: processing episode: {current_episode + 1} / {expert_data_num}", end="\r")
+
+        load_path = os.path.join(load_data_dir, f"data/episode_{current_episode:07d}.hdf5")
+        data = load_hdf5(load_path)
+        
+        state_all = pack_robot_state(data, action_type, robot_action_dim_info, source_type="dataset", state_type="state")
+        action_all = pack_robot_state(data, action_type, robot_action_dim_info, source_type="dataset", state_type="action")
+
+        for j in range(0, state_all.shape[0]):
+            head_img_bit = data['vision']['cam_head']['colors'][j]
+            left_img_bit = data['vision']['cam_left_wrist']['colors'][j]
+            right_img_bit = data['vision']['cam_right_wrist']['colors'][j]
+
+            state, action = state_all[j], action_all[j]
+
+            head_img = decode_image_bit(head_img_bit)
+            assert head_img.ndim == 3 and head_img.shape[-1] == 3, f"Expected HxWx3, got {head_img.shape}"
+            head_img = cv2.resize(head_img, (320, 240), interpolation=cv2.INTER_AREA)  # (W, H)
+            assert head_img.shape == (240, 320, 3)
+
+            left_img = decode_image_bit(left_img_bit)
+            assert left_img.ndim == 3 and left_img.shape[-1] == 3, f"Expected HxWx3, got {left_img.shape}"
+            left_img = cv2.resize(left_img, (320, 240), interpolation=cv2.INTER_AREA)  # (W, H)
+            assert left_img.shape == (240, 320, 3)
+
+            right_img = decode_image_bit(right_img_bit)
+            assert right_img.ndim == 3 and right_img.shape[-1] == 3, f"Expected HxWx3, got {right_img.shape}"
+            right_img = cv2.resize(right_img, (320, 240), interpolation=cv2.INTER_AREA)  # (W, H)
+            assert right_img.shape == (240, 320, 3)
+
+            head_camera_arrays.append(head_img)
+            left_camera_arrays.append(left_img)
+            right_camera_arrays.append(right_img)
+            state_arrays.append(state)
+            action_arrays.append(action)
+
+        current_episode += 1
+        frame_count += state_all.shape[0]
+        episode_ends_arrays.append(frame_count)
+
+    print()
+    episode_ends_arrays = np.array(episode_ends_arrays)
+    action_arrays = np.array(action_arrays)
+    state_arrays = np.array(state_arrays)
+
+    head_camera_arrays = np.array(head_camera_arrays)
+    head_camera_arrays = np.moveaxis(head_camera_arrays, -1, 1)  # NHWC -> NCHW
+
+    left_camera_arrays = np.array(left_camera_arrays)
+    left_camera_arrays = np.moveaxis(left_camera_arrays, -1, 1)  # NHWC -> NCHW
+
+    right_camera_arrays = np.array(right_camera_arrays)
+    right_camera_arrays = np.moveaxis(right_camera_arrays, -1, 1)  # NHWC -> NCHW
+
+    compressor = zarr.Blosc(cname="zstd", clevel=3, shuffle=1)
+    action_chunk_size = (100, action_arrays.shape[1])
+    state_chunk_size = (100, state_arrays.shape[1])
+    head_camera_chunk_size = (100, *head_camera_arrays.shape[1:])
+    left_camera_chunk_size = (100, *left_camera_arrays.shape[1:])
+    right_camera_chunk_size = (100, *right_camera_arrays.shape[1:])
+
+    zarr_data.create_dataset("head_camera", data=head_camera_arrays, chunks=head_camera_chunk_size, overwrite=True, compressor=compressor,)
+    zarr_data.create_dataset("left_camera", data=left_camera_arrays, chunks=left_camera_chunk_size, overwrite=True, compressor=compressor,)
+    zarr_data.create_dataset("right_camera", data=right_camera_arrays, chunks=right_camera_chunk_size, overwrite=True, compressor=compressor,)
+    zarr_data.create_dataset("state", data=state_arrays, chunks=state_chunk_size, dtype="float32", overwrite=True, compressor=compressor,)
+    zarr_data.create_dataset("action", data=action_arrays, chunks=action_chunk_size,dtype="float32", overwrite=True, compressor=compressor,)
+    zarr_meta.create_dataset("episode_ends", data=episode_ends_arrays, dtype="int64", overwrite=True, compressor=compressor,)
+
+if __name__ == "__main__":
+    main()

real_inference/source/XPolicyLab/policy/Pi_05/openpi/packages/openpi-client/pyproject.toml

@@ -0,0 +1,23 @@
+[project]
+name = "openpi-client"
+version = "0.1.0"
+requires-python = ">=3.7"
+dependencies = [
+    "dm-tree>=0.1.8",
+    "msgpack>=1.0.5",
+    "numpy>=1.22.4,<2.0.0",
+    "pillow>=9.0.0",
+    "tree>=0.2.4",
+    "websockets>=11.0",
+]
+
+[build-system]
+requires = ["hatchling"]
+build-backend = "hatchling.build"
+
+[tool.uv]
+dev-dependencies = ["pytest>=8.3.4"]
+
+[tool.ruff]
+line-length = 120
+target-version = "py37"

real_inference/source/XPolicyLab/policy/Pi_05/openpi/packages/openpi-client/src/openpi_client/__init__.py

@@ -0,0 +1 @@
+__version__ = "0.1.0"

real_inference/source/XPolicyLab/policy/Pi_05/openpi/packages/openpi-client/src/openpi_client/action_chunk_broker.py

@@ -0,0 +1,50 @@
+from typing import Dict
+
+import numpy as np
+import tree
+from typing_extensions import override
+
+from openpi_client import base_policy as _base_policy
+
+
+class ActionChunkBroker(_base_policy.BasePolicy):
+    """Wraps a policy to return action chunks one-at-a-time.
+
+    Assumes that the first dimension of all action fields is the chunk size.
+
+    A new inference call to the inner policy is only made when the current
+    list of chunks is exhausted.
+    """
+
+    def __init__(self, policy: _base_policy.BasePolicy, action_horizon: int):
+        self._policy = policy
+        self._action_horizon = action_horizon
+        self._cur_step: int = 0
+
+        self._last_results: Dict[str, np.ndarray] | None = None
+
+    @override
+    def infer(self, obs: Dict) -> Dict:  # noqa: UP006
+        if self._last_results is None:
+            self._last_results = self._policy.infer(obs)
+            self._cur_step = 0
+
+        def slicer(x):
+            if isinstance(x, np.ndarray):
+                return x[self._cur_step, ...]
+            else:
+                return x
+
+        results = tree.map_structure(slicer, self._last_results)
+        self._cur_step += 1
+
+        if self._cur_step >= self._action_horizon:
+            self._last_results = None
+
+        return results
+
+    @override
+    def reset(self) -> None:
+        self._policy.reset()
+        self._last_results = None
+        self._cur_step = 0

real_inference/source/XPolicyLab/policy/Pi_05/openpi/packages/openpi-client/src/openpi_client/base_policy.py

@@ -0,0 +1,12 @@
+import abc
+from typing import Dict
+
+
+class BasePolicy(abc.ABC):
+    @abc.abstractmethod
+    def infer(self, obs: Dict) -> Dict:
+        """Infer actions from observations."""
+
+    def reset(self) -> None:
+        """Reset the policy to its initial state."""
+        pass

real_inference/source/XPolicyLab/policy/Pi_05/openpi/packages/openpi-client/src/openpi_client/image_tools.py

@@ -0,0 +1,58 @@
+import numpy as np
+from PIL import Image
+
+
+def convert_to_uint8(img: np.ndarray) -> np.ndarray:
+    """Converts an image to uint8 if it is a float image.
+
+    This is important for reducing the size of the image when sending it over the network.
+    """
+    if np.issubdtype(img.dtype, np.floating):
+        img = (255 * img).astype(np.uint8)
+    return img
+
+
+def resize_with_pad(images: np.ndarray, height: int, width: int, method=Image.BILINEAR) -> np.ndarray:
+    """Replicates tf.image.resize_with_pad for multiple images using PIL. Resizes a batch of images to a target height.
+
+    Args:
+        images: A batch of images in [..., height, width, channel] format.
+        height: The target height of the image.
+        width: The target width of the image.
+        method: The interpolation method to use. Default is bilinear.
+
+    Returns:
+        The resized images in [..., height, width, channel].
+    """
+    # If the images are already the correct size, return them as is.
+    if images.shape[-3:-1] == (height, width):
+        return images
+
+    original_shape = images.shape
+
+    images = images.reshape(-1, *original_shape[-3:])
+    resized = np.stack([_resize_with_pad_pil(Image.fromarray(im), height, width, method=method) for im in images])
+    return resized.reshape(*original_shape[:-3], *resized.shape[-3:])
+
+
+def _resize_with_pad_pil(image: Image.Image, height: int, width: int, method: int) -> Image.Image:
+    """Replicates tf.image.resize_with_pad for one image using PIL. Resizes an image to a target height and
+    width without distortion by padding with zeros.
+
+    Unlike the jax version, note that PIL uses [width, height, channel] ordering instead of [batch, h, w, c].
+    """
+    cur_width, cur_height = image.size
+    if cur_width == width and cur_height == height:
+        return image  # No need to resize if the image is already the correct size.
+
+    ratio = max(cur_width / width, cur_height / height)
+    resized_height = int(cur_height / ratio)
+    resized_width = int(cur_width / ratio)
+    resized_image = image.resize((resized_width, resized_height), resample=method)
+
+    zero_image = Image.new(resized_image.mode, (width, height), 0)
+    pad_height = max(0, int((height - resized_height) / 2))
+    pad_width = max(0, int((width - resized_width) / 2))
+    zero_image.paste(resized_image, (pad_width, pad_height))
+    assert zero_image.size == (width, height)
+    return zero_image

real_inference/source/XPolicyLab/policy/Pi_05/openpi/packages/openpi-client/src/openpi_client/image_tools_test.py

@@ -0,0 +1,37 @@
+import numpy as np
+
+import openpi_client.image_tools as image_tools
+
+
+def test_resize_with_pad_shapes():
+    # Test case 1: Resize image with larger dimensions
+    images = np.zeros((2, 10, 10, 3), dtype=np.uint8)  # Input images of shape (batch_size, height, width, channels)
+    height = 20
+    width = 20
+    resized_images = image_tools.resize_with_pad(images, height, width)
+    assert resized_images.shape == (2, height, width, 3)
+    assert np.all(resized_images == 0)
+
+    # Test case 2: Resize image with smaller dimensions
+    images = np.zeros((3, 30, 30, 3), dtype=np.uint8)
+    height = 15
+    width = 15
+    resized_images = image_tools.resize_with_pad(images, height, width)
+    assert resized_images.shape == (3, height, width, 3)
+    assert np.all(resized_images == 0)
+
+    # Test case 3: Resize image with the same dimensions
+    images = np.zeros((1, 50, 50, 3), dtype=np.uint8)
+    height = 50
+    width = 50
+    resized_images = image_tools.resize_with_pad(images, height, width)
+    assert resized_images.shape == (1, height, width, 3)
+    assert np.all(resized_images == 0)
+
+    # Test case 3: Resize image with odd-numbered padding
+    images = np.zeros((1, 256, 320, 3), dtype=np.uint8)
+    height = 60
+    width = 80
+    resized_images = image_tools.resize_with_pad(images, height, width)
+    assert resized_images.shape == (1, height, width, 3)
+    assert np.all(resized_images == 0)

real_inference/source/XPolicyLab/policy/Pi_05/openpi/packages/openpi-client/src/openpi_client/msgpack_numpy.py

@@ -0,0 +1,57 @@
+"""Adds NumPy array support to msgpack.
+
+msgpack is good for (de)serializing data over a network for multiple reasons:
+- msgpack is secure (as opposed to pickle/dill/etc which allow for arbitrary code execution)
+- msgpack is widely used and has good cross-language support
+- msgpack does not require a schema (as opposed to protobuf/flatbuffers/etc) which is convenient in dynamically typed
+    languages like Python and JavaScript
+- msgpack is fast and efficient (as opposed to readable formats like JSON/YAML/etc); I found that msgpack was ~4x faster
+    than pickle for serializing large arrays using the below strategy
+
+The code below is adapted from https://github.com/lebedov/msgpack-numpy. The reason not to use that library directly is
+that it falls back to pickle for object arrays.
+"""
+
+import functools
+
+import msgpack
+import numpy as np
+
+
+def pack_array(obj):
+    if (isinstance(obj, (np.ndarray, np.generic))) and obj.dtype.kind in ("V", "O", "c"):
+        raise ValueError(f"Unsupported dtype: {obj.dtype}")
+
+    if isinstance(obj, np.ndarray):
+        return {
+            b"__ndarray__": True,
+            b"data": obj.tobytes(),
+            b"dtype": obj.dtype.str,
+            b"shape": obj.shape,
+        }
+
+    if isinstance(obj, np.generic):
+        return {
+            b"__npgeneric__": True,
+            b"data": obj.item(),
+            b"dtype": obj.dtype.str,
+        }
+
+    return obj
+
+
+def unpack_array(obj):
+    if b"__ndarray__" in obj:
+        return np.ndarray(buffer=obj[b"data"], dtype=np.dtype(obj[b"dtype"]), shape=obj[b"shape"])
+
+    if b"__npgeneric__" in obj:
+        return np.dtype(obj[b"dtype"]).type(obj[b"data"])
+
+    return obj
+
+
+Packer = functools.partial(msgpack.Packer, default=pack_array)
+packb = functools.partial(msgpack.packb, default=pack_array)
+
+Unpacker = functools.partial(msgpack.Unpacker, object_hook=unpack_array)
+unpackb = functools.partial(msgpack.unpackb, object_hook=unpack_array)

real_inference/source/XPolicyLab/policy/Pi_05/openpi/packages/openpi-client/src/openpi_client/msgpack_numpy_test.py

@@ -0,0 +1,45 @@
+import numpy as np
+import pytest
+import tree
+
+from openpi_client import msgpack_numpy
+
+
+def _check(expected, actual):
+    if isinstance(expected, np.ndarray):
+        assert expected.shape == actual.shape
+        assert expected.dtype == actual.dtype
+        assert np.array_equal(expected, actual, equal_nan=expected.dtype.kind == "f")
+    else:
+        assert expected == actual
+
+
+@pytest.mark.parametrize(
+    "data",
+    [
+        1,  # int
+        1.0,  # float
+        "hello",  # string
+        np.bool_(True),  # boolean scalar
+        np.array([1, 2, 3])[0],  # int scalar
+        np.str_("asdf"),  # string scalar
+        [1, 2, 3],  # list
+        {"key": "value"},  # dict
+        {"key": [1, 2, 3]},  # nested dict
+        np.array(1.0),  # 0D array
+        np.array([1, 2, 3], dtype=np.int32),  # 1D integer array
+        np.array(["asdf", "qwer"]),  # string array
+        np.array([True, False]),  # boolean array
+        np.array([[1.0, 2.0], [3.0, 4.0]], dtype=np.float32),  # 2D float array
+        np.array([[[1, 2], [3, 4]], [[5, 6], [7, 8]]], dtype=np.int16),  # 3D integer array
+        np.array([np.nan, np.inf, -np.inf]),  # special float values
+        {"arr": np.array([1, 2, 3]), "nested": {"arr": np.array([4, 5, 6])}},  # nested dict with arrays
+        [np.array([1, 2]), np.array([3, 4])],  # list of arrays
+        np.zeros((3, 4, 5), dtype=np.float32),  # 3D zeros
+        np.ones((2, 3), dtype=np.float64),  # 2D ones with double precision
+    ],
+)
+def test_pack_unpack(data):
+    packed = msgpack_numpy.packb(data)
+    unpacked = msgpack_numpy.unpackb(packed)
+    tree.map_structure(_check, data, unpacked)

real_inference/source/XPolicyLab/policy/Pi_05/openpi/packages/openpi-client/src/openpi_client/runtime/agent.py

@@ -0,0 +1,17 @@
+import abc
+
+
+class Agent(abc.ABC):
+    """An Agent is the thing with agency, i.e. the entity that makes decisions.
+
+    Agents receive observations about the state of the world, and return actions
+    to take in response.
+    """
+
+    @abc.abstractmethod
+    def get_action(self, observation: dict) -> dict:
+        """Query the agent for the next action."""
+
+    @abc.abstractmethod
+    def reset(self) -> None:
+        """Reset the agent to its initial state."""

real_inference/source/XPolicyLab/policy/Pi_05/openpi/packages/openpi-client/src/openpi_client/runtime/agents/policy_agent.py

@@ -0,0 +1,18 @@
+from typing_extensions import override
+
+from openpi_client import base_policy as _base_policy
+from openpi_client.runtime import agent as _agent
+
+
+class PolicyAgent(_agent.Agent):
+    """An agent that uses a policy to determine actions."""
+
+    def __init__(self, policy: _base_policy.BasePolicy) -> None:
+        self._policy = policy
+
+    @override
+    def get_action(self, observation: dict) -> dict:
+        return self._policy.infer(observation)
+
+    def reset(self) -> None:
+        self._policy.reset()

real_inference/source/XPolicyLab/policy/Pi_05/openpi/packages/openpi-client/src/openpi_client/runtime/environment.py

@@ -0,0 +1,32 @@
+import abc
+
+
+class Environment(abc.ABC):
+    """An Environment represents the robot and the environment it inhabits.
+
+    The primary contract of environments is that they can be queried for observations
+    about their state, and have actions applied to them to change that state.
+    """
+
+    @abc.abstractmethod
+    def reset(self) -> None:
+        """Reset the environment to its initial state.
+
+        This will be called once before starting each episode.
+        """
+
+    @abc.abstractmethod
+    def is_episode_complete(self) -> bool:
+        """Allow the environment to signal that the episode is complete.
+
+        This will be called after each step. It should return `True` if the episode is
+        complete (either successfully or unsuccessfully), and `False` otherwise.
+        """
+
+    @abc.abstractmethod
+    def get_observation(self) -> dict:
+        """Query the environment for the current state."""
+
+    @abc.abstractmethod
+    def apply_action(self, action: dict) -> None:
+        """Take an action in the environment."""

real_inference/source/XPolicyLab/policy/Pi_05/openpi/packages/openpi-client/src/openpi_client/runtime/runtime.py

@@ -0,0 +1,92 @@
+import logging
+import threading
+import time
+
+from openpi_client.runtime import agent as _agent
+from openpi_client.runtime import environment as _environment
+from openpi_client.runtime import subscriber as _subscriber
+
+
+class Runtime:
+    """The core module orchestrating interactions between key components of the system."""
+
+    def __init__(
+        self,
+        environment: _environment.Environment,
+        agent: _agent.Agent,
+        subscribers: list[_subscriber.Subscriber],
+        max_hz: float = 0,
+        num_episodes: int = 1,
+        max_episode_steps: int = 0,
+    ) -> None:
+        self._environment = environment
+        self._agent = agent
+        self._subscribers = subscribers
+        self._max_hz = max_hz
+        self._num_episodes = num_episodes
+        self._max_episode_steps = max_episode_steps
+
+        self._in_episode = False
+        self._episode_steps = 0
+
+    def run(self) -> None:
+        """Runs the runtime loop continuously until stop() is called or the environment is done."""
+        for _ in range(self._num_episodes):
+            self._run_episode()
+
+        # Final reset, this is important for real environments to move the robot to its home position.
+        self._environment.reset()
+
+    def run_in_new_thread(self) -> threading.Thread:
+        """Runs the runtime loop in a new thread."""
+        thread = threading.Thread(target=self.run)
+        thread.start()
+        return thread
+
+    def mark_episode_complete(self) -> None:
+        """Marks the end of an episode."""
+        self._in_episode = False
+
+    def _run_episode(self) -> None:
+        """Runs a single episode."""
+        logging.info("Starting episode...")
+        self._environment.reset()
+        self._agent.reset()
+        for subscriber in self._subscribers:
+            subscriber.on_episode_start()
+
+        self._in_episode = True
+        self._episode_steps = 0
+        step_time = 1 / self._max_hz if self._max_hz > 0 else 0
+        last_step_time = time.time()
+
+        while self._in_episode:
+            self._step()
+            self._episode_steps += 1
+
+            # Sleep to maintain the desired frame rate
+            now = time.time()
+            dt = now - last_step_time
+            if dt < step_time:
+                time.sleep(step_time - dt)
+                last_step_time = time.time()
+            else:
+                last_step_time = now
+
+        logging.info("Episode completed.")
+        for subscriber in self._subscribers:
+            subscriber.on_episode_end()
+
+    def _step(self) -> None:
+        """A single step of the runtime loop."""
+        observation = self._environment.get_observation()
+        action = self._agent.get_action(observation)
+        self._environment.apply_action(action)
+
+        for subscriber in self._subscribers:
+            subscriber.on_step(observation, action)
+
+        if self._environment.is_episode_complete() or (
+            self._max_episode_steps > 0 and self._episode_steps >= self._max_episode_steps
+        ):
+            self.mark_episode_complete()

real_inference/source/XPolicyLab/policy/Pi_05/openpi/packages/openpi-client/src/openpi_client/runtime/subscriber.py

@@ -0,0 +1,20 @@
+import abc
+
+
+class Subscriber(abc.ABC):
+    """Subscribes to events in the runtime.
+
+    Subscribers can be used to save data, visualize, etc.
+    """
+
+    @abc.abstractmethod
+    def on_episode_start(self) -> None:
+        """Called when an episode starts."""
+
+    @abc.abstractmethod
+    def on_step(self, observation: dict, action: dict) -> None:
+        """Append a step to the episode."""
+
+    @abc.abstractmethod
+    def on_episode_end(self) -> None:
+        """Called when an episode ends."""

real_inference/source/XPolicyLab/policy/Pi_05/openpi/packages/openpi-client/src/openpi_client/websocket_client_policy.py

@@ -0,0 +1,58 @@
+import logging
+import time
+from typing import Dict, Optional, Tuple
+
+from typing_extensions import override
+import websockets.sync.client
+
+from openpi_client import base_policy as _base_policy
+from openpi_client import msgpack_numpy
+
+
+class WebsocketClientPolicy(_base_policy.BasePolicy):
+    """Implements the Policy interface by communicating with a server over websocket.
+
+    See WebsocketPolicyServer for a corresponding server implementation.
+    """
+
+    def __init__(self, host: str = "0.0.0.0", port: Optional[int] = None, api_key: Optional[str] = None) -> None:
+        if host.startswith("ws"):
+            self._uri = host
+        else:
+            self._uri = f"ws://{host}"
+        if port is not None:
+            self._uri += f":{port}"
+        self._packer = msgpack_numpy.Packer()
+        self._api_key = api_key
+        self._ws, self._server_metadata = self._wait_for_server()
+
+    def get_server_metadata(self) -> Dict:
+        return self._server_metadata
+
+    def _wait_for_server(self) -> Tuple[websockets.sync.client.ClientConnection, Dict]:
+        logging.info(f"Waiting for server at {self._uri}...")
+        while True:
+            try:
+                headers = {"Authorization": f"Api-Key {self._api_key}"} if self._api_key else None
+                conn = websockets.sync.client.connect(
+                    self._uri, compression=None, max_size=None, additional_headers=headers
+                )
+                metadata = msgpack_numpy.unpackb(conn.recv())
+                return conn, metadata
+            except ConnectionRefusedError:
+                logging.info("Still waiting for server...")
+                time.sleep(5)
+
+    @override
+    def infer(self, obs: Dict) -> Dict:  # noqa: UP006
+        data = self._packer.pack(obs)
+        self._ws.send(data)
+        response = self._ws.recv()
+        if isinstance(response, str):
+            # we're expecting bytes; if the server sends a string, it's an error.
+            raise RuntimeError(f"Error in inference server:\n{response}")
+        return msgpack_numpy.unpackb(response)
+
+    @override
+    def reset(self) -> None:
+        pass

real_inference/source/XPolicyLab/policy/Pi_05/openpi/src/openpi/models/model_test.py

@@ -0,0 +1,94 @@
+from flax import nnx
+import jax
+import pytest
+
+from openpi.models import model as _model
+from openpi.models import pi0_config
+from openpi.models import pi0_fast
+from openpi.shared import download
+from openpi.shared import nnx_utils
+
+
+def test_pi0_model():
+    key = jax.random.key(0)
+    config = pi0_config.Pi0Config()
+    model = config.create(key)
+
+    batch_size = 2
+    obs, act = config.fake_obs(batch_size), config.fake_act(batch_size)
+
+    loss = nnx_utils.module_jit(model.compute_loss)(key, obs, act)
+    assert loss.shape == (batch_size, config.action_horizon)
+
+    actions = nnx_utils.module_jit(model.sample_actions)(key, obs, num_steps=10)
+    assert actions.shape == (batch_size, model.action_horizon, model.action_dim)
+
+
+def test_pi0_lora_model():
+    key = jax.random.key(0)
+    config = pi0_config.Pi0Config(paligemma_variant="gemma_2b_lora")
+    model = config.create(key)
+
+    batch_size = 2
+    obs, act = config.fake_obs(batch_size), config.fake_act(batch_size)
+
+    loss = nnx_utils.module_jit(model.compute_loss)(key, obs, act)
+    assert loss.shape == (batch_size, config.action_horizon)
+
+    actions = nnx_utils.module_jit(model.sample_actions)(key, obs, num_steps=10)
+    assert actions.shape == (batch_size, model.action_horizon, model.action_dim)
+
+
+def test_pi0_fast_model():
+    key = jax.random.key(0)
+    config = pi0_fast.Pi0FASTConfig()
+    model = config.create(key)
+
+    batch_size = 2
+    obs, act = config.fake_obs(batch_size), config.fake_act(batch_size)
+
+    loss = nnx_utils.module_jit(model.compute_loss)(key, obs, act)
+    assert loss.shape == (batch_size,)
+
+    actions = nnx_utils.module_jit(model.sample_actions)(key, obs)
+    assert actions.shape == (batch_size, 256)
+
+
+def test_pi0_fast_lora_model():
+    key = jax.random.key(0)
+    config = pi0_fast.Pi0FASTConfig(paligemma_variant="gemma_2b_lora")
+    model = config.create(key)
+
+    batch_size = 2
+    obs, act = config.fake_obs(batch_size), config.fake_act(batch_size)
+
+    loss = nnx_utils.module_jit(model.compute_loss)(key, obs, act)
+    assert loss.shape == (batch_size,)
+
+    actions = nnx_utils.module_jit(model.sample_actions)(key, obs)
+    assert actions.shape == (batch_size, 256)
+
+    lora_filter = nnx_utils.PathRegex(".*lora.*")
+    model_state = nnx.state(model)
+
+    lora_state_elems = list(model_state.filter(lora_filter))
+    assert len(lora_state_elems) > 0
+
+
+@pytest.mark.manual
+def test_model_restore():
+    key = jax.random.key(0)
+    config = pi0_config.Pi0Config()
+
+    batch_size = 2
+    obs, act = config.fake_obs(batch_size), config.fake_act(batch_size)
+
+    model = config.load(
+        _model.restore_params(download.maybe_download("gs://openpi-assets/checkpoints/pi0_base/params"))
+    )
+
+    loss = model.compute_loss(key, obs, act)
+    assert loss.shape == (batch_size, config.action_horizon)
+
+    actions = model.sample_actions(key, obs, num_steps=10)
+    assert actions.shape == (batch_size, model.action_horizon, model.action_dim)

real_inference/source/XPolicyLab/policy/Pi_05/openpi/src/openpi/models/pi0.py

@@ -0,0 +1,279 @@
+import logging
+
+import einops
+import flax.nnx as nnx
+import flax.nnx.bridge as nnx_bridge
+import jax
+import jax.numpy as jnp
+from typing_extensions import override
+
+from openpi.models import model as _model
+from openpi.models import pi0_config
+import openpi.models.gemma as _gemma
+import openpi.models.siglip as _siglip
+from openpi.shared import array_typing as at
+
+logger = logging.getLogger("openpi")
+
+
+def make_attn_mask(input_mask, mask_ar):
+    """Adapted from big_vision.
+
+    Tokens can attend to valid inputs tokens which have a cumulative mask_ar
+    smaller or equal to theirs. This way `mask_ar` bool[?B, N] can be used to
+    setup several types of attention, for example:
+
+      [[1 1 1 1 1 1]]: pure causal attention.
+
+      [[0 0 0 1 1 1]]: prefix-lm attention. The first 3 tokens can attend between
+          themselves and the last 3 tokens have a causal attention. The first
+          entry could also be a 1 without changing behaviour.
+
+      [[1 0 1 0 1 0 0 1 0 0]]: causal attention between 4 blocks. Tokens of a
+          block can attend all previous blocks and all tokens on the same block.
+
+    Args:
+      input_mask: bool[B, N] true if its part of the input, false if padding.
+      mask_ar: bool[?B, N] mask that's true where previous tokens cannot depend on
+        it and false where it shares the same attention mask as the previous token.
+    """
+    mask_ar = jnp.broadcast_to(mask_ar, input_mask.shape)
+    cumsum = jnp.cumsum(mask_ar, axis=1)
+    attn_mask = cumsum[:, None, :] <= cumsum[:, :, None]
+    valid_mask = input_mask[:, None, :] * input_mask[:, :, None]
+    return jnp.logical_and(attn_mask, valid_mask)
+
+
+@at.typecheck
+def posemb_sincos(
+    pos: at.Real[at.Array, " b"], embedding_dim: int, min_period: float, max_period: float
+) -> at.Float[at.Array, "b {embedding_dim}"]:
+    """Computes sine-cosine positional embedding vectors for scalar positions."""
+    if embedding_dim % 2 != 0:
+        raise ValueError(f"embedding_dim ({embedding_dim}) must be divisible by 2")
+
+    fraction = jnp.linspace(0.0, 1.0, embedding_dim // 2)
+    period = min_period * (max_period / min_period) ** fraction
+    sinusoid_input = jnp.einsum(
+        "i,j->ij",
+        pos,
+        1.0 / period * 2 * jnp.pi,
+        precision=jax.lax.Precision.HIGHEST,
+    )
+    return jnp.concatenate([jnp.sin(sinusoid_input), jnp.cos(sinusoid_input)], axis=-1)
+
+
+class Pi0(_model.BaseModel):
+    def __init__(self, config: pi0_config.Pi0Config, rngs: nnx.Rngs):
+        super().__init__(config.action_dim, config.action_horizon, config.max_token_len)
+        self.pi05 = config.pi05
+        paligemma_config = _gemma.get_config(config.paligemma_variant)
+        action_expert_config = _gemma.get_config(config.action_expert_variant)
+        # TODO: rewrite gemma in NNX. For now, use bridge.
+        llm = nnx_bridge.ToNNX(
+            _gemma.Module(
+                configs=[paligemma_config, action_expert_config],
+                embed_dtype=config.dtype,
+                adarms=config.pi05,
+            )
+        )
+        llm.lazy_init(rngs=rngs, method="init", use_adarms=[False, True] if config.pi05 else [False, False])
+        img = nnx_bridge.ToNNX(
+            _siglip.Module(
+                num_classes=paligemma_config.width,
+                variant="So400m/14",
+                pool_type="none",
+                scan=True,
+                dtype_mm=config.dtype,
+            )
+        )
+        img.lazy_init(next(iter(config.fake_obs().images.values())), train=False, rngs=rngs)
+        self.PaliGemma = nnx.Dict(llm=llm, img=img)
+        self.action_in_proj = nnx.Linear(config.action_dim, action_expert_config.width, rngs=rngs)
+        if config.pi05:
+            self.time_mlp_in = nnx.Linear(action_expert_config.width, action_expert_config.width, rngs=rngs)
+            self.time_mlp_out = nnx.Linear(action_expert_config.width, action_expert_config.width, rngs=rngs)
+        else:
+            self.state_proj = nnx.Linear(config.action_dim, action_expert_config.width, rngs=rngs)
+            self.action_time_mlp_in = nnx.Linear(2 * action_expert_config.width, action_expert_config.width, rngs=rngs)
+            self.action_time_mlp_out = nnx.Linear(action_expert_config.width, action_expert_config.width, rngs=rngs)
+        self.action_out_proj = nnx.Linear(action_expert_config.width, config.action_dim, rngs=rngs)
+
+        # This attribute gets automatically set by model.train() and model.eval().
+        self.deterministic = True
+
+    @at.typecheck
+    def embed_prefix(
+        self, obs: _model.Observation
+    ) -> tuple[at.Float[at.Array, "b s emb"], at.Bool[at.Array, "b s"], at.Bool[at.Array, " s"]]:
+        input_mask = []
+        ar_mask = []
+        tokens = []
+        # embed images
+        for name in obs.images:
+            image_tokens, _ = self.PaliGemma.img(obs.images[name], train=False)
+
+            tokens.append(image_tokens)
+            input_mask.append(
+                einops.repeat(
+                    obs.image_masks[name],
+                    "b -> b s",
+                    s=image_tokens.shape[1],
+                )
+            )
+            # image tokens attend to each other
+            ar_mask += [False] * image_tokens.shape[1]
+
+        # add language (aka tokenized inputs)
+        if obs.tokenized_prompt is not None:
+            tokenized_inputs = self.PaliGemma.llm(obs.tokenized_prompt, method="embed")
+            tokens.append(tokenized_inputs)
+            input_mask.append(obs.tokenized_prompt_mask)
+            # full attention between image and language inputs
+            ar_mask += [False] * tokenized_inputs.shape[1]
+        tokens = jnp.concatenate(tokens, axis=1)
+        input_mask = jnp.concatenate(input_mask, axis=1)
+        ar_mask = jnp.array(ar_mask)
+        return tokens, input_mask, ar_mask
+
+    @at.typecheck
+    def embed_suffix(
+        self, obs: _model.Observation, noisy_actions: _model.Actions, timestep: at.Float[at.Array, " b"]
+    ) -> tuple[
+        at.Float[at.Array, "b s emb"],
+        at.Bool[at.Array, "b s"],
+        at.Bool[at.Array, " s"],
+        at.Float[at.Array, "b emb"] | None,
+    ]:
+        input_mask = []
+        ar_mask = []
+        tokens = []
+        if not self.pi05:
+            # add a single state token
+            state_token = self.state_proj(obs.state)[:, None, :]
+            tokens.append(state_token)
+            input_mask.append(jnp.ones((obs.state.shape[0], 1), dtype=jnp.bool_))
+            # image/language inputs do not attend to state or actions
+            ar_mask += [True]
+
+        action_tokens = self.action_in_proj(noisy_actions)
+        # embed timestep using sine-cosine positional encoding with sensitivity in the range [0, 1]
+        time_emb = posemb_sincos(timestep, self.action_in_proj.out_features, min_period=4e-3, max_period=4.0)
+        if self.pi05:
+            # time MLP (for adaRMS)
+            time_emb = self.time_mlp_in(time_emb)
+            time_emb = nnx.swish(time_emb)
+            time_emb = self.time_mlp_out(time_emb)
+            time_emb = nnx.swish(time_emb)
+            action_expert_tokens = action_tokens
+            adarms_cond = time_emb
+        else:
+            # mix timestep + action information using an MLP (no adaRMS)
+            time_tokens = einops.repeat(time_emb, "b emb -> b s emb", s=self.action_horizon)
+            action_time_tokens = jnp.concatenate([action_tokens, time_tokens], axis=-1)
+            action_time_tokens = self.action_time_mlp_in(action_time_tokens)
+            action_time_tokens = nnx.swish(action_time_tokens)
+            action_time_tokens = self.action_time_mlp_out(action_time_tokens)
+            action_expert_tokens = action_time_tokens
+            adarms_cond = None
+        tokens.append(action_expert_tokens)
+        input_mask.append(jnp.ones(action_expert_tokens.shape[:2], dtype=jnp.bool_))
+        # image/language/state inputs do not attend to action tokens
+        ar_mask += [True] + ([False] * (self.action_horizon - 1))
+        tokens = jnp.concatenate(tokens, axis=1)
+        input_mask = jnp.concatenate(input_mask, axis=1)
+        ar_mask = jnp.array(ar_mask)
+        return tokens, input_mask, ar_mask, adarms_cond
+
+    @override
+    def compute_loss(
+        self, rng: at.KeyArrayLike, observation: _model.Observation, actions: _model.Actions, *, train: bool = False
+    ) -> at.Float[at.Array, "*b ah"]:
+        preprocess_rng, noise_rng, time_rng = jax.random.split(rng, 3)
+        observation = _model.preprocess_observation(preprocess_rng, observation, train=train)
+
+        batch_shape = actions.shape[:-2]
+        noise = jax.random.normal(noise_rng, actions.shape)
+        time = jax.random.beta(time_rng, 1.5, 1, batch_shape) * 0.999 + 0.001
+        time_expanded = time[..., None, None]
+        x_t = time_expanded * noise + (1 - time_expanded) * actions
+        u_t = noise - actions
+
+        # one big forward pass of prefix + suffix at once
+        prefix_tokens, prefix_mask, prefix_ar_mask = self.embed_prefix(observation)
+        suffix_tokens, suffix_mask, suffix_ar_mask, adarms_cond = self.embed_suffix(observation, x_t, time)
+        input_mask = jnp.concatenate([prefix_mask, suffix_mask], axis=1)
+        ar_mask = jnp.concatenate([prefix_ar_mask, suffix_ar_mask], axis=0)
+        attn_mask = make_attn_mask(input_mask, ar_mask)
+        positions = jnp.cumsum(input_mask, axis=1) - 1
+        (prefix_out, suffix_out), _ = self.PaliGemma.llm(
+            [prefix_tokens, suffix_tokens], mask=attn_mask, positions=positions, adarms_cond=[None, adarms_cond]
+        )
+        v_t = self.action_out_proj(suffix_out[:, -self.action_horizon :])
+
+        return jnp.mean(jnp.square(v_t - u_t), axis=-1)
+
+    @override
+    def sample_actions(
+        self,
+        rng: at.KeyArrayLike,
+        observation: _model.Observation,
+        *,
+        num_steps: int | at.Int[at.Array, ""] = 10,
+        noise: at.Float[at.Array, "b ah ad"] | None = None,
+    ) -> _model.Actions:
+        observation = _model.preprocess_observation(None, observation, train=False)
+        # note that we use the convention more common in diffusion literature, where t=1 is noise and t=0 is the target
+        # distribution. yes, this is the opposite of the pi0 paper, and I'm sorry.
+        dt = -1.0 / num_steps
+        batch_size = observation.state.shape[0]
+        if noise is None:
+            noise = jax.random.normal(rng, (batch_size, self.action_horizon, self.action_dim))
+
+        # first fill KV cache with a forward pass of the prefix
+        prefix_tokens, prefix_mask, prefix_ar_mask = self.embed_prefix(observation)
+        prefix_attn_mask = make_attn_mask(prefix_mask, prefix_ar_mask)
+        positions = jnp.cumsum(prefix_mask, axis=1) - 1
+        _, kv_cache = self.PaliGemma.llm([prefix_tokens, None], mask=prefix_attn_mask, positions=positions)
+
+        def step(carry):
+            x_t, time = carry
+            suffix_tokens, suffix_mask, suffix_ar_mask, adarms_cond = self.embed_suffix(
+                observation, x_t, jnp.broadcast_to(time, batch_size)
+            )
+            # `suffix_attn_mask` is shape (b, suffix_len, suffix_len) indicating how the suffix tokens can attend to each
+            # other
+            suffix_attn_mask = make_attn_mask(suffix_mask, suffix_ar_mask)
+            # `prefix_attn_mask` is shape (b, suffix_len, prefix_len) indicating how the suffix tokens can attend to the
+            # prefix tokens
+            prefix_attn_mask = einops.repeat(prefix_mask, "b p -> b s p", s=suffix_tokens.shape[1])
+            # `combined_mask` is shape (b, suffix_len, prefix_len + suffix_len) indicating how the suffix tokens (which
+            # generate the queries) can attend to the full prefix + suffix sequence (which generates the keys and values)
+            full_attn_mask = jnp.concatenate([prefix_attn_mask, suffix_attn_mask], axis=-1)
+            assert full_attn_mask.shape == (
+                batch_size,
+                suffix_tokens.shape[1],
+                prefix_tokens.shape[1] + suffix_tokens.shape[1],
+            )
+            # `positions` is shape (b, suffix_len) indicating the positions of the suffix tokens
+            positions = jnp.sum(prefix_mask, axis=-1)[:, None] + jnp.cumsum(suffix_mask, axis=-1) - 1
+
+            (prefix_out, suffix_out), _ = self.PaliGemma.llm(
+                [None, suffix_tokens],
+                mask=full_attn_mask,
+                positions=positions,
+                kv_cache=kv_cache,
+                adarms_cond=[None, adarms_cond],
+            )
+            assert prefix_out is None
+            v_t = self.action_out_proj(suffix_out[:, -self.action_horizon :])
+
+            return x_t + dt * v_t, time + dt
+
+        def cond(carry):
+            x_t, time = carry
+            # robust to floating-point error
+            return time >= -dt / 2
+
+        x_0, _ = jax.lax.while_loop(cond, step, (noise, 1.0))
+        return x_0

real_inference/source/XPolicyLab/policy/Pi_05/openpi/src/openpi/models/pi0_config.py

@@ -0,0 +1,117 @@
+import dataclasses
+from typing import TYPE_CHECKING
+
+import flax.nnx as nnx
+import jax
+import jax.numpy as jnp
+from typing_extensions import override
+
+from openpi.models import model as _model
+import openpi.models.gemma as _gemma
+from openpi.shared import array_typing as at
+import openpi.shared.nnx_utils as nnx_utils
+
+if TYPE_CHECKING:
+    from openpi.models.pi0 import Pi0
+
+
+@dataclasses.dataclass(frozen=True)
+class Pi0Config(_model.BaseModelConfig):
+    dtype: str = "bfloat16"
+    paligemma_variant: _gemma.Variant = "gemma_2b"
+    action_expert_variant: _gemma.Variant = "gemma_300m"
+
+    # Set the model specific defaults.
+    action_dim: int = 32
+    action_horizon: int = 50
+    max_token_len: int = None  # type: ignore
+    # Pi05 has two differences from Pi0:
+    # - the state input is part of the discrete language tokens rather than a continuous input that is part of the suffix
+    # - the action expert uses adaRMSNorm to inject the flow matching timestep
+    pi05: bool = False
+    # This config option is not used directly by the model, but it is read by the ModelTransformFactory.
+    discrete_state_input: bool = None  # type: ignore
+
+    pytorch_compile_mode: str | None = "max-autotune"
+
+    def __post_init__(self):
+        if self.max_token_len is None:
+            object.__setattr__(self, "max_token_len", 200 if self.pi05 else 48)
+        if self.discrete_state_input is None:
+            object.__setattr__(self, "discrete_state_input", self.pi05)
+        if self.pytorch_compile_mode is not None:
+            assert self.pytorch_compile_mode in [
+                "default",
+                "reduce-overhead",
+                "max-autotune",
+                "max-autotune-no-cudagraphs",
+            ]
+
+    @property
+    @override
+    def model_type(self) -> _model.ModelType:
+        if self.pi05:
+            return _model.ModelType.PI05
+        return _model.ModelType.PI0
+
+    @override
+    def create(self, rng: at.KeyArrayLike) -> "Pi0":
+        from openpi.models.pi0 import Pi0
+
+        return Pi0(self, rngs=nnx.Rngs(rng))
+
+    @override
+    def inputs_spec(self, *, batch_size: int = 1) -> tuple[_model.Observation, _model.Actions]:
+        image_spec = jax.ShapeDtypeStruct([batch_size, *_model.IMAGE_RESOLUTION, 3], jnp.float32)
+        image_mask_spec = jax.ShapeDtypeStruct([batch_size], jnp.bool_)
+
+        with at.disable_typechecking():
+            observation_spec = _model.Observation(
+                images={
+                    "base_0_rgb": image_spec,
+                    "left_wrist_0_rgb": image_spec,
+                    "right_wrist_0_rgb": image_spec,
+                },
+                image_masks={
+                    "base_0_rgb": image_mask_spec,
+                    "left_wrist_0_rgb": image_mask_spec,
+                    "right_wrist_0_rgb": image_mask_spec,
+                },
+                state=jax.ShapeDtypeStruct([batch_size, self.action_dim], jnp.float32),
+                tokenized_prompt=jax.ShapeDtypeStruct([batch_size, self.max_token_len], jnp.int32),
+                tokenized_prompt_mask=jax.ShapeDtypeStruct([batch_size, self.max_token_len], bool),
+            )
+        action_spec = jax.ShapeDtypeStruct([batch_size, self.action_horizon, self.action_dim], jnp.float32)
+
+        return observation_spec, action_spec
+
+    def get_freeze_filter(self) -> nnx.filterlib.Filter:
+        """Returns the freeze filter based on the model config."""
+        filters = []
+        has_lora = False
+        gemma_params_filter = nnx_utils.PathRegex(".*llm.*")
+        action_expert_params_filter = nnx_utils.PathRegex(".*llm.*_1.*")
+        if "lora" in self.paligemma_variant:
+            filters.append(
+                gemma_params_filter,
+            )
+            if "lora" not in self.action_expert_variant:
+                # If only freeze gemma params, exclude action expert params.
+                filters.append(
+                    nnx.Not(action_expert_params_filter),
+                )
+            has_lora = True
+        elif "lora" in self.action_expert_variant:
+            filters.append(
+                action_expert_params_filter,
+            )
+            has_lora = True
+
+        if has_lora:
+            # If any lora is used, exclude all lora params.
+            filters.append(
+                nnx.Not(nnx_utils.PathRegex(".*lora.*")),
+            )
+        if not filters:
+            return nnx.Nothing
+        return nnx.All(*filters)

real_inference/source/XPolicyLab/policy/Pi_05/openpi/src/openpi/models/utils/fsq_tokenizer.py

@@ -0,0 +1,472 @@
+import math
+from typing import Any, Literal
+
+import chex
+from einops import einops
+from flax import linen as nn
+from flax.linen.module import Module
+from flax.linen.module import compact
+from flax.struct import dataclass
+from flax.typing import Array
+import jax
+import jax.numpy as jnp
+
+
+class FsqCodebook(nn.Module):
+    input_dim: int
+    target_codebook_size: int
+    codebook_type: Literal["fsq", "lfq"]
+
+    _bins_per_dim: tuple[int] | None = None
+
+    @property
+    def bins_per_dim(self) -> tuple[int]:
+        if self._bins_per_dim is not None:
+            return self._bins_per_dim
+
+        if self.codebook_type == "fsq":
+            return self._get_bins_fsq(self.target_codebook_size)
+        elif self.codebook_type == "lfq":  # noqa: RET505
+            return self._get_bins_lfq(self.target_codebook_size)
+        elif self.codebook_type == "custom":
+            return self._get_bins_custom(self.target_codebook_size)
+        else:
+            raise ValueError(f"Codebook type {self.codebook_type} not supported.")
+
+    @property
+    def place_values(self) -> jnp.ndarray:
+        place_values = [1]
+        for b in self.bins_per_dim[:-1]:
+            place_values.append(place_values[-1] * b)
+        return jnp.array(place_values)
+
+    @staticmethod
+    def _get_bins_fsq(target_codebook_size: int) -> tuple[int]:
+        """
+        Get bins per dimension based on codebook size, from the original FSQ paper.
+        """
+        if target_codebook_size == 2**8:
+            return (8, 6, 5)
+        elif target_codebook_size == 2**10:  # noqa: RET505
+            return (8, 5, 5, 5)
+        elif target_codebook_size == 2**12:
+            return (7, 5, 5, 5, 5)
+        elif target_codebook_size == 2**14:
+            return (8, 8, 8, 6, 5)
+        elif target_codebook_size == 2**16:
+            return (8, 8, 8, 5, 5, 5)
+        else:
+            raise ValueError(f"Codebook size {target_codebook_size} not supported.")
+
+    @staticmethod
+    def _get_bins_custom(target_codebook_size: int) -> tuple[int]:
+        if target_codebook_size == 2**8:
+            return (16, 16)
+        elif target_codebook_size == 2**10:  # noqa: RET505
+            return (32, 32)
+        elif target_codebook_size == 2**12:
+            return (64, 64)
+        elif target_codebook_size == 2**14:
+            return (128, 128)
+        elif target_codebook_size == 2**16:
+            return (256, 256)
+        return None
+
+    @staticmethod
+    def _get_bins_lfq(target_codebook_size: int) -> tuple[int]:
+        """
+        Get bins per dimension according to the Lookup-Free Quantization paper (2 bins per dimension)
+        """
+        assert target_codebook_size & (target_codebook_size - 1) == 0, "Codebook size should be a power of two for LFQ"
+
+        return (2,) * int(math.log2(target_codebook_size))
+
+    def setup(self):
+        self.proj_down = nn.Dense(len(self.bins_per_dim))
+        self.proj_up = nn.Dense(self.input_dim)
+
+    def __call__(self, inputs: jnp.ndarray) -> tuple[jnp.ndarray, jnp.ndarray]:
+        tokens, z = self.encode(inputs)
+        output = self.decode(tokens, z_grad=z)
+        return tokens, output
+
+    def encode(self, inputs: jnp.ndarray) -> tuple[jnp.ndarray, jnp.ndarray]:
+        bases = jnp.array(self.bins_per_dim)
+
+        x = self.proj_down(inputs)
+        z = jnp.tanh(x)
+
+        # Quantize
+        digits = jnp.round((z + 1) * (bases - 1) / 2).astype(jnp.int32)
+        tokens = self.undigitize(digits)
+
+        return tokens, z
+
+    def decode(self, tokens: jnp.ndarray, z_grad: jax.Array | None = None) -> jnp.ndarray:
+        bases = jnp.array(self.bins_per_dim)
+        digits = self.digitize(tokens)
+
+        z_q = digits / (bases - 1) * 2 - 1
+
+        if z_grad is not None:
+            chex.assert_equal_shape([z_q, z_grad])
+            z_q = jax.lax.stop_gradient(z_q - z_grad) + z_grad
+
+        return self.proj_up(z_q)
+
+    def undigitize(self, digits: jnp.ndarray) -> jnp.ndarray:
+        return jnp.sum(digits * jnp.array(self.place_values), axis=-1)
+
+    def digitize(self, tokens: jnp.ndarray) -> jnp.ndarray:
+        return (tokens[..., None] // jnp.array(self.place_values)) % jnp.array(self.bins_per_dim)
+
+    @property
+    def vocab_size(self) -> int:
+        return math.prod(self.bins_per_dim)
+
+
+class ResNetDownBlock(nn.Module):
+    stride: int = 1
+    n_filters: int = 64
+    dropout_rate: float = 0.0
+    group_size: int = 32
+
+    @nn.compact
+    def __call__(self, x: jnp.ndarray, *, train: bool = True) -> jnp.ndarray:
+        skip = x
+
+        if self.stride > 1 or x.shape[-1] != self.n_filters:
+            skip = nn.Conv(self.n_filters, (self.stride,), (self.stride,), "SAME")(skip)
+
+        x = nn.Conv(self.n_filters, (3,), (self.stride,), "SAME")(x)
+        x = nn.GroupNorm(num_groups=self.n_filters // self.group_size)(x)
+        x = nn.Dropout(self.dropout_rate)(x, deterministic=not train)
+        x = nn.relu(x)
+        x = nn.Conv(self.n_filters, (3,), (1,), "SAME")(x)
+
+        return skip + x
+
+
+class ResNetUpBlock(nn.Module):
+    stride: int = 1
+    n_filters: int = 64
+    dropout_rate: float = 0.0
+    group_size: int = 32
+
+    @nn.compact
+    def __call__(self, x: jnp.ndarray, *, train: bool = True) -> jnp.ndarray:
+        skip = x
+
+        if self.stride > 1:
+            skip = nn.ConvTranspose(self.n_filters, (self.stride,), (self.stride,), "SAME")(skip)
+
+        x = nn.ConvTranspose(self.n_filters, (3,), (self.stride,), "SAME")(x)
+        x = nn.GroupNorm(num_groups=self.n_filters // self.group_size)(x)
+        x = nn.Dropout(self.dropout_rate)(x, deterministic=not train)
+        x = nn.relu(x)
+        x = nn.ConvTranspose(self.n_filters, (3,), (1,), "SAME")(x)
+
+        return skip + x
+
+
+@dataclass
+class LfqCodebookOutput:
+    tokens: jnp.ndarray
+    z: jnp.ndarray
+    z_q: jnp.ndarray
+    token_log_probs: jnp.ndarray
+    commit_loss: jnp.ndarray
+
+
+class LookupFreeQuantization(nn.Module):
+    num_dims: int
+    latent_dim: int
+
+    def setup(self):
+        self.codebook = jnp.array([-1, 1])
+        self.activation = nn.tanh
+
+        self.project_down = nn.Dense(self.num_dims)
+        self.project_up = nn.Dense(self.latent_dim)
+
+    def encode(self, z: jnp.ndarray) -> jnp.ndarray:
+        z = self.project_down(z)
+        token_squared_distances = jnp.square(z[..., None] - self.codebook)
+        token_bits = jnp.argmin(token_squared_distances, axis=-1)
+        return jnp.sum(token_bits * (2 ** jnp.arange(self.num_dims)), axis=-1)
+
+    def decode(self, tokens: jnp.ndarray) -> jnp.ndarray:
+        token_bits = (tokens[..., None] & (2 ** jnp.arange(self.num_dims))).astype(jnp.int32)
+        return self.project_up(self.codebook[token_bits])
+
+    def loss(self, x: jnp.ndarray) -> LfqCodebookOutput:
+        z = self.project_down(x)
+        z = self.activation(z)
+
+        token_squared_distances = jnp.square(z[..., None] - self.codebook)
+        tokens = jnp.argmin(token_squared_distances, axis=-1)
+
+        token_bit_log_probs = -token_squared_distances
+        # Compute token log probs for tokens 0..2^num_dims-1 by summing corresponding log-probs
+        token_bit_expansions = jnp.bitwise_and(
+            jnp.arange(2**self.num_dims)[None, :], 2 ** jnp.arange(self.num_dims)[:, None]
+        ).astype(jnp.int32)
+        token_log_probs = (
+            token_bit_log_probs[..., 0] @ (1 - token_bit_expansions)
+            + token_bit_log_probs[..., 1] @ token_bit_expansions
+        )  # (batch_size, num_tokens, 2 ** num_dims)
+        token_log_probs = jax.lax.stop_gradient(jax.nn.log_softmax(token_log_probs, axis=-1))
+        chex.assert_shape(token_log_probs, (*x.shape[:-1], 2**self.num_dims))
+
+        z_q = self.codebook[tokens]
+        commit_loss = jnp.square(z - z_q).mean()
+        z_q = jax.lax.stop_gradient(z_q - z) + z
+
+        z_q = self.project_up(z_q)
+        z = self.project_up(z)
+
+        tokens = jnp.sum(tokens * (len(self.codebook) ** jnp.arange(self.num_dims)), axis=-1)
+        return LfqCodebookOutput(
+            tokens=tokens,
+            z=z,
+            z_q=z_q,
+            token_log_probs=jnp.zeros(()),
+            commit_loss=commit_loss,
+        )
+
+
+def make_block_causal_attention_matrix(q: jnp.ndarray, k: jnp.ndarray, bs_q: int, bs_k: int) -> jnp.ndarray:
+    return nn.make_attention_mask(q, k, pairwise_fn=lambda x, y: jnp.greater_equal(x // bs_k, y // bs_q))
+
+
+class GeGLU(Module):
+    """Gated Linear Unit with GELU (GeGLU) activation function.
+    GeGLU is a Flax layer that combines a linear transformation with a GELU
+    activation function in a gating mechanism. It is often used in Transformer models
+    to provide non-linear capabilities while preserving a strong linear component.
+
+    Attributes:
+        features: the number of output features (default: None).
+    """
+
+    output_dim: int = -1
+
+    @compact
+    def __call__(self, inputs: Array) -> Array:
+        """Applies the GeGLU activation to the inputs.
+        Args:
+            inputs: the nd-array to apply the GeGLU activation function to.
+        Returns:
+            The transformed input.
+        """
+        output_dim = inputs.shape[-1] if self.output_dim == -1 else self.output_dim
+
+        x = nn.Dense(output_dim * 2)(inputs)
+        x, gate = x[..., :output_dim], x[..., output_dim:]
+        return x * nn.gelu(gate)
+
+
+class CrossAttentionLayer(nn.Module):
+    dropout_rate: float = 0.0
+    num_heads: int = None
+    causal: bool = False
+    mlp_ratio: float = 4.0
+
+    @nn.compact
+    def __call__(
+        self,
+        x: jnp.ndarray,
+        y: jnp.ndarray,
+        *,
+        mask_self: jnp.ndarray | None = None,
+        mask_cross: jnp.ndarray | None = None,
+        train: bool = True,
+    ) -> jnp.ndarray:
+        d_embed = x.shape[-1]
+        seq_len_q = x.shape[-2]
+        seq_len_k = y.shape[-2]
+
+        if self.causal:
+            # One block size will be 1
+            bs_q = max(seq_len_q // seq_len_k, 1)
+            bs_k = max(seq_len_k // seq_len_q, 1)
+
+            mask_self = nn.make_causal_mask(x[..., 0])
+            mask_cross = make_block_causal_attention_matrix(x[..., 0], y[..., 0], bs_q, bs_k)
+
+        # Self-attention block
+        skip = x
+        x = nn.LayerNorm()(x)
+        x = nn.MultiHeadDotProductAttention(
+            num_heads=self.num_heads or d_embed // 64,
+            dropout_rate=self.dropout_rate,
+            deterministic=not train,
+        )(x, x, x, mask=mask_self)
+        x = skip + x
+
+        # Cross-attention block
+        skip = x
+        x = nn.LayerNorm()(x)
+        x = nn.MultiHeadDotProductAttention(
+            num_heads=self.num_heads or d_embed // 64,
+            dropout_rate=self.dropout_rate,
+            deterministic=not train,
+        )(x, y, y, mask=mask_cross)
+        x = skip + x
+
+        # MLP block
+        skip = x
+        x = nn.LayerNorm()(x)
+        x = nn.Dense(int(d_embed * self.mlp_ratio))(x)
+        x = nn.Dropout(self.dropout_rate)(x, deterministic=not train)
+        x = GeGLU()(x)
+        x = nn.Dense(d_embed)(x)
+        return skip + x
+
+
+def sinusoidal_pe_init(_, shape: tuple[int, int]) -> jnp.ndarray:
+    seq_len, d_embed = shape
+
+    position = jnp.arange(0, seq_len, 1)
+    div_term = jnp.exp(jnp.arange(0, d_embed, 2) * -(jnp.log(10000.0) / d_embed))
+    return jnp.concatenate(
+        [
+            jnp.sin(position[:, jnp.newaxis] * div_term),
+            jnp.cos(position[:, jnp.newaxis] * div_term),
+        ],
+        axis=-1,
+    )
+
+
+class TokenizerEncoderDecoder(nn.Module):
+    num_tokens: int
+    num_cross_tokens: int
+    num_layers: int
+    causal: bool
+
+    mlp_ratio: float = 4.0
+    use_state_conditioning: bool = False
+
+    @nn.compact
+    def __call__(
+        self,
+        y: jnp.ndarray,
+        *,
+        train: bool = True,
+        state_conditioning: jnp.ndarray | None = None,
+        mask: jnp.ndarray | None = None,
+    ) -> jnp.ndarray:
+        x = self.param("q_embed", sinusoidal_pe_init, (self.num_tokens, y.shape[-1]))
+        x = jax.numpy.broadcast_to(x, y.shape[:-2] + x.shape[-2:])
+
+        if mask is not None:
+            # mask is (batch_dims..., num_cross_tokens)
+            chex.assert_equal_shape([y[..., 0], mask])
+            attn_mask = einops.repeat(mask, "... kv -> ... 1 q kv", q=self.num_tokens)
+        else:
+            attn_mask = jnp.ones((*y.shape[:-2], 1, self.num_tokens, self.num_cross_tokens))
+
+        if self.use_state_conditioning:
+            assert state_conditioning is not None, "State conditioning is required for this model."
+            state_embed = nn.Dense(y.shape[-1], name="state_proj")(state_conditioning)[..., None, :]
+            y = jnp.concatenate([y, state_embed], axis=-2)
+            attn_mask = jnp.concatenate([attn_mask, jnp.ones_like(attn_mask[..., 0:1])], axis=-1)
+
+        y = y + self.param("y_pos_enc", sinusoidal_pe_init, y.shape[-2:])
+
+        for _ in range(self.num_layers):
+            x = CrossAttentionLayer(causal=self.causal, mlp_ratio=self.mlp_ratio)(
+                x, y, train=train, mask_self=None, mask_cross=attn_mask
+            )
+
+        return x
+
+
+class FsqAttentionTokenizer(nn.Module):
+    embed_dim: int
+    data_dim: int
+    data_horizon: int
+    num_tokens: int
+    num_layers: int
+    target_codebook_size: int
+    causal: bool = False
+    mlp_ratio: float = 2.0
+
+    bound: float | None = None
+
+    use_state_conditioning: bool = False
+
+    @property
+    def vocab_size(self) -> int:
+        return math.prod(FsqCodebook._get_bins_fsq(self.target_codebook_size))  # noqa: SLF001
+
+    def setup(self):
+        self.proj = nn.Dense(self.embed_dim)
+        self.encoder = TokenizerEncoderDecoder(
+            num_tokens=self.num_tokens,
+            num_cross_tokens=self.data_horizon,
+            num_layers=self.num_layers,
+            causal=self.causal,
+            use_state_conditioning=self.use_state_conditioning,
+            mlp_ratio=self.mlp_ratio,
+        )
+        self.codebook = FsqCodebook(
+            input_dim=self.embed_dim,
+            target_codebook_size=self.target_codebook_size,
+            codebook_type="custom",
+        )
+        self.decoder = TokenizerEncoderDecoder(
+            num_tokens=self.data_horizon,
+            num_cross_tokens=self.num_tokens,
+            num_layers=self.num_layers,
+            causal=self.causal,
+            use_state_conditioning=self.use_state_conditioning,
+            mlp_ratio=self.mlp_ratio,
+        )
+
+        self.proj_mean = nn.Dense(self.data_dim)
+        self.out_scale = self.param("out_scale", lambda _: jnp.full((), 1.0))
+
+    def tokenize(
+        self, action: jnp.ndarray, *, obs: jnp.ndarray | None = None, train: bool = False
+    ) -> tuple[jnp.ndarray, jnp.ndarray]:
+        if self.bound is not None:
+            action = jnp.clip(action, -self.bound, self.bound)
+
+        x = self.proj(action)
+        x = self.encoder(x, train=train, state_conditioning=obs)
+
+        return self.codebook.encode(x)
+
+    def detokenize(self, tokens: jnp.ndarray, *, obs: jnp.ndarray | None = None) -> jnp.ndarray:
+        x = self.decoder(self.codebook.decode(tokens), state_conditioning=obs)
+        mean = self.proj_mean(x)
+        return mean * self.out_scale
+
+    def loss(
+        self, action: jnp.ndarray, *, obs: jnp.ndarray | None = None, train: bool = True
+    ) -> tuple[jnp.ndarray, dict[str, jnp.ndarray]]:
+        # Encode
+        x = self.proj(action)
+        z = self.encoder(x, train=train, state_conditioning=obs)
+
+        # Quantize
+        tokens, z = self.codebook(z)
+
+        # Decode
+        x = self.decoder(z, train=train, state_conditioning=obs)
+        mean = self.proj_mean(x) * self.out_scale
+
+        mse = jnp.mean(jnp.square(action - mean))
+        mae = jnp.mean(jnp.abs(action - mean))
+
+        return mse, {
+            "mse": mse,
+            "mae": mae,
+        }
+
+    def __call__(self, *args: Any, **kwargs: Any) -> tuple[jnp.ndarray, dict[str, jnp.ndarray]]:
+        """
+        Dummy for .init
+        """
+        return self.loss(*args, **kwargs)

real_inference/source/XPolicyLab/policy/Pi_05/openpi/src/openpi/models_pytorch/gemma_pytorch.py

@@ -0,0 +1,281 @@
+from typing import Literal
+
+import pytest
+import torch
+from torch import nn
+from transformers import GemmaForCausalLM
+from transformers import PaliGemmaForConditionalGeneration
+from transformers.models.auto import CONFIG_MAPPING
+from transformers.models.gemma import modeling_gemma
+
+
+class PaliGemmaWithExpertModel(nn.Module):
+    def __init__(
+        self,
+        vlm_config,
+        action_expert_config,
+        use_adarms=None,
+        precision: Literal["bfloat16", "float32"] = "bfloat16",
+    ):
+        if use_adarms is None:
+            use_adarms = [False, False]
+        super().__init__()
+
+        vlm_config_hf = CONFIG_MAPPING["paligemma"]()
+        vlm_config_hf._vocab_size = 257152  # noqa: SLF001
+        vlm_config_hf.image_token_index = 257152
+        vlm_config_hf.text_config.hidden_size = vlm_config.width
+        vlm_config_hf.text_config.intermediate_size = vlm_config.mlp_dim
+        vlm_config_hf.text_config.num_attention_heads = vlm_config.num_heads
+        vlm_config_hf.text_config.head_dim = vlm_config.head_dim
+        vlm_config_hf.text_config.num_hidden_layers = vlm_config.depth
+        vlm_config_hf.text_config.num_key_value_heads = vlm_config.num_kv_heads
+        vlm_config_hf.text_config.hidden_activation = "gelu_pytorch_tanh"
+        vlm_config_hf.text_config.torch_dtype = "float32"
+        vlm_config_hf.text_config.vocab_size = 257152
+        vlm_config_hf.text_config.use_adarms = use_adarms[0]
+        vlm_config_hf.text_config.adarms_cond_dim = vlm_config.width if use_adarms[0] else None
+        vlm_config_hf.vision_config.intermediate_size = 4304
+        vlm_config_hf.vision_config.projection_dim = 2048
+        vlm_config_hf.vision_config.projector_hidden_act = "gelu_fast"
+        vlm_config_hf.vision_config.torch_dtype = "float32"
+
+        action_expert_config_hf = CONFIG_MAPPING["gemma"](
+            head_dim=action_expert_config.head_dim,
+            hidden_size=action_expert_config.width,
+            intermediate_size=action_expert_config.mlp_dim,
+            num_attention_heads=action_expert_config.num_heads,
+            num_hidden_layers=action_expert_config.depth,
+            num_key_value_heads=action_expert_config.num_kv_heads,
+            vocab_size=257152,
+            hidden_activation="gelu_pytorch_tanh",
+            torch_dtype="float32",
+            use_adarms=use_adarms[1],
+            adarms_cond_dim=action_expert_config.width if use_adarms[1] else None,
+        )
+
+        self.paligemma = PaliGemmaForConditionalGeneration(config=vlm_config_hf)
+        self.gemma_expert = GemmaForCausalLM(config=action_expert_config_hf)
+        self.gemma_expert.model.embed_tokens = None
+
+        self.to_bfloat16_for_selected_params(precision)
+
+    def to_bfloat16_for_selected_params(self, precision: Literal["bfloat16", "float32"] = "bfloat16"):
+        if precision == "bfloat16":
+            self.to(dtype=torch.bfloat16)
+        elif precision == "float32":
+            self.to(dtype=torch.float32)
+            return
+        else:
+            raise ValueError(f"Invalid precision: {precision}")
+
+        params_to_keep_float32 = [
+            "vision_tower.vision_model.embeddings.patch_embedding.weight",
+            "vision_tower.vision_model.embeddings.patch_embedding.bias",
+            "vision_tower.vision_model.embeddings.position_embedding.weight",
+            "input_layernorm",
+            "post_attention_layernorm",
+            "model.norm",
+        ]
+
+        for name, param in self.named_parameters():
+            if any(selector in name for selector in params_to_keep_float32):
+                param.data = param.data.to(dtype=torch.float32)
+
+    def embed_image(self, image: torch.Tensor):
+        return self.paligemma.model.get_image_features(image)
+
+    def embed_language_tokens(self, tokens: torch.Tensor):
+        return self.paligemma.language_model.embed_tokens(tokens)
+
+    def forward(
+        self,
+        attention_mask: torch.Tensor | None = None,
+        position_ids: torch.LongTensor | None = None,
+        past_key_values: list[torch.FloatTensor] | pytest.Cache | None = None,
+        inputs_embeds: list[torch.FloatTensor] | None = None,
+        use_cache: bool | None = None,
+        adarms_cond: list[torch.Tensor] | None = None,
+    ):
+        if adarms_cond is None:
+            adarms_cond = [None, None]
+        if inputs_embeds[1] is None:
+            prefix_output = self.paligemma.language_model.forward(
+                inputs_embeds=inputs_embeds[0],
+                attention_mask=attention_mask,
+                position_ids=position_ids,
+                past_key_values=past_key_values,
+                use_cache=use_cache,
+                adarms_cond=adarms_cond[0] if adarms_cond is not None else None,
+            )
+            prefix_past_key_values = prefix_output.past_key_values
+            prefix_output = prefix_output.last_hidden_state
+            suffix_output = None
+        elif inputs_embeds[0] is None:
+            suffix_output = self.gemma_expert.model.forward(
+                inputs_embeds=inputs_embeds[1],
+                attention_mask=attention_mask,
+                position_ids=position_ids,
+                past_key_values=past_key_values,
+                use_cache=use_cache,
+                adarms_cond=adarms_cond[1] if adarms_cond is not None else None,
+            )
+            suffix_output = suffix_output.last_hidden_state
+            prefix_output = None
+            prefix_past_key_values = None
+        else:
+            models = [self.paligemma.language_model, self.gemma_expert.model]
+            num_layers = self.paligemma.config.text_config.num_hidden_layers
+
+            # Check if gradient checkpointing is enabled for any of the models
+            use_gradient_checkpointing = (
+                hasattr(self.gemma_expert.model, "gradient_checkpointing")
+                and self.gemma_expert.model.gradient_checkpointing
+                and self.training
+            ) or (hasattr(self, "gradient_checkpointing") and self.gradient_checkpointing and self.training)
+
+            # Force enable gradient checkpointing if we're in training mode and the model supports it
+            if self.training and hasattr(self.gemma_expert.model, "gradient_checkpointing"):
+                if not self.gemma_expert.model.gradient_checkpointing:
+                    print("Forcing gradient checkpointing to be enabled for Gemma expert model")
+                    self.gemma_expert.model.gradient_checkpointing = True
+                use_gradient_checkpointing = True
+
+            # Debug gradient checkpointing status
+            if hasattr(self, "_debug_gc_printed") and not self._debug_gc_printed:
+                print(f"Gemma expert model gradient checkpointing: {use_gradient_checkpointing}")
+                print(f"Model training mode: {self.training}")
+                print(
+                    f"Gemma expert model has gradient_checkpointing attr: {hasattr(self.gemma_expert.model, 'gradient_checkpointing')}"
+                )
+                if hasattr(self.gemma_expert.model, "gradient_checkpointing"):
+                    print(
+                        f"Gemma expert model gradient_checkpointing value: {self.gemma_expert.model.gradient_checkpointing}"
+                    )
+                self._debug_gc_printed = True
+
+            # Define the complete layer computation function for gradient checkpointing
+            def compute_layer_complete(layer_idx, inputs_embeds, attention_mask, position_ids, adarms_cond):
+                models = [self.paligemma.language_model, self.gemma_expert.model]
+
+                query_states = []
+                key_states = []
+                value_states = []
+                gates = []
+                for i, hidden_states in enumerate(inputs_embeds):
+                    layer = models[i].layers[layer_idx]
+                    hidden_states, gate = layer.input_layernorm(hidden_states, cond=adarms_cond[i])  # noqa: PLW2901
+                    gates.append(gate)
+
+                    input_shape = hidden_states.shape[:-1]
+                    hidden_shape = (*input_shape, -1, layer.self_attn.head_dim)
+                    query_state = layer.self_attn.q_proj(hidden_states).view(hidden_shape).transpose(1, 2)
+                    key_state = layer.self_attn.k_proj(hidden_states).view(hidden_shape).transpose(1, 2)
+                    value_state = layer.self_attn.v_proj(hidden_states).view(hidden_shape).transpose(1, 2)
+
+                    query_states.append(query_state)
+                    key_states.append(key_state)
+                    value_states.append(value_state)
+
+                # Concatenate and process attention
+                query_states = torch.cat(query_states, dim=2)
+                key_states = torch.cat(key_states, dim=2)
+                value_states = torch.cat(value_states, dim=2)
+
+                dummy_tensor = torch.zeros(
+                    query_states.shape[0],
+                    query_states.shape[2],
+                    query_states.shape[-1],
+                    device=query_states.device,
+                    dtype=query_states.dtype,
+                )
+                cos, sin = self.paligemma.model.language_model.rotary_emb(dummy_tensor, position_ids)
+                query_states, key_states = modeling_gemma.apply_rotary_pos_emb(
+                    query_states, key_states, cos, sin, unsqueeze_dim=1
+                )
+
+                batch_size = query_states.shape[0]
+                scaling = self.paligemma.language_model.layers[layer_idx].self_attn.scaling
+
+                # Attention computation
+                att_output, _ = modeling_gemma.eager_attention_forward(
+                    self.paligemma.language_model.layers[layer_idx].self_attn,
+                    query_states,
+                    key_states,
+                    value_states,
+                    attention_mask,
+                    scaling,
+                )
+                # Get head_dim from the current layer, not from the model
+                head_dim = self.paligemma.language_model.layers[layer_idx].self_attn.head_dim
+                att_output = att_output.reshape(batch_size, -1, 1 * 8 * head_dim)
+
+                # Process layer outputs
+                outputs_embeds = []
+                start_pos = 0
+                for i, hidden_states in enumerate(inputs_embeds):
+                    layer = models[i].layers[layer_idx]
+                    end_pos = start_pos + hidden_states.shape[1]
+
+                    if att_output.dtype != layer.self_attn.o_proj.weight.dtype:
+                        att_output = att_output.to(layer.self_attn.o_proj.weight.dtype)
+                    out_emb = layer.self_attn.o_proj(att_output[:, start_pos:end_pos])
+
+                    # first residual
+                    out_emb = modeling_gemma._gated_residual(hidden_states, out_emb, gates[i])  # noqa: SLF001
+                    after_first_residual = out_emb.clone()
+                    out_emb, gate = layer.post_attention_layernorm(out_emb, cond=adarms_cond[i])
+                    # Convert to bfloat16 if the next layer (mlp) uses bfloat16
+                    if layer.mlp.up_proj.weight.dtype == torch.bfloat16:
+                        out_emb = out_emb.to(dtype=torch.bfloat16)
+
+                    out_emb = layer.mlp(out_emb)
+                    # second residual
+                    out_emb = modeling_gemma._gated_residual(after_first_residual, out_emb, gate)  # noqa: SLF001
+                    outputs_embeds.append(out_emb)
+                    start_pos = end_pos
+
+                return outputs_embeds
+
+            # Process all layers with gradient checkpointing if enabled
+            for layer_idx in range(num_layers):
+                if use_gradient_checkpointing:
+                    inputs_embeds = torch.utils.checkpoint.checkpoint(
+                        compute_layer_complete,
+                        layer_idx,
+                        inputs_embeds,
+                        attention_mask,
+                        position_ids,
+                        adarms_cond,
+                        use_reentrant=False,
+                        preserve_rng_state=False,
+                    )
+                else:
+                    inputs_embeds = compute_layer_complete(
+                        layer_idx, inputs_embeds, attention_mask, position_ids, adarms_cond
+                    )
+
+                # Old code removed - now using compute_layer_complete function above
+
+            # final norm
+            # Define final norm computation function for gradient checkpointing
+            def compute_final_norms(inputs_embeds, adarms_cond):
+                outputs_embeds = []
+                for i, hidden_states in enumerate(inputs_embeds):
+                    out_emb, _ = models[i].norm(hidden_states, cond=adarms_cond[i])
+                    outputs_embeds.append(out_emb)
+                return outputs_embeds
+
+            # Apply gradient checkpointing to final norm if enabled
+            if use_gradient_checkpointing:
+                outputs_embeds = torch.utils.checkpoint.checkpoint(
+                    compute_final_norms, inputs_embeds, adarms_cond, use_reentrant=False, preserve_rng_state=False
+                )
+            else:
+                outputs_embeds = compute_final_norms(inputs_embeds, adarms_cond)
+
+            prefix_output = outputs_embeds[0]
+            suffix_output = outputs_embeds[1]
+            prefix_past_key_values = None
+
+        return [prefix_output, suffix_output], prefix_past_key_values

real_inference/source/XPolicyLab/policy/Pi_05/openpi/src/openpi/models_pytorch/pi0_pytorch.py

@@ -0,0 +1,462 @@
+import logging
+import math
+
+import torch
+from torch import Tensor
+from torch import nn
+import torch.nn.functional as F  # noqa: N812
+
+import openpi.models.gemma as _gemma
+from openpi.models_pytorch.gemma_pytorch import PaliGemmaWithExpertModel
+import openpi.models_pytorch.preprocessing_pytorch as _preprocessing
+
+
+def get_safe_dtype(target_dtype, device_type):
+    """Get a safe dtype for the given device type."""
+    if device_type == "cpu":
+        # CPU doesn't support bfloat16, use float32 instead
+        if target_dtype == torch.bfloat16:
+            return torch.float32
+        if target_dtype == torch.float64:
+            return torch.float64
+    return target_dtype
+
+
+def create_sinusoidal_pos_embedding(
+    time: torch.tensor, dimension: int, min_period: float, max_period: float, device="cpu"
+) -> Tensor:
+    """Computes sine-cosine positional embedding vectors for scalar positions."""
+    if dimension % 2 != 0:
+        raise ValueError(f"dimension ({dimension}) must be divisible by 2")
+
+    if time.ndim != 1:
+        raise ValueError("The time tensor is expected to be of shape `(batch_size, )`.")
+
+    dtype = get_safe_dtype(torch.float64, device.type)
+    fraction = torch.linspace(0.0, 1.0, dimension // 2, dtype=dtype, device=device)
+    period = min_period * (max_period / min_period) ** fraction
+
+    # Compute the outer product
+    scaling_factor = 1.0 / period * 2 * math.pi
+    sin_input = scaling_factor[None, :] * time[:, None]
+    return torch.cat([torch.sin(sin_input), torch.cos(sin_input)], dim=1)
+
+
+def sample_beta(alpha, beta, bsize, device):
+    alpha_t = torch.as_tensor(alpha, dtype=torch.float32, device=device)
+    beta_t = torch.as_tensor(beta, dtype=torch.float32, device=device)
+    dist = torch.distributions.Beta(alpha_t, beta_t)
+    return dist.sample((bsize,))
+
+
+def make_att_2d_masks(pad_masks, att_masks):
+    """Copied from big_vision.
+
+    Tokens can attend to valid inputs tokens which have a cumulative mask_ar
+    smaller or equal to theirs. This way `mask_ar` int[B, N] can be used to
+    setup several types of attention, for example:
+
+      [[1 1 1 1 1 1]]: pure causal attention.
+
+      [[0 0 0 1 1 1]]: prefix-lm attention. The first 3 tokens can attend between
+          themselves and the last 3 tokens have a causal attention. The first
+          entry could also be a 1 without changing behaviour.
+
+      [[1 0 1 0 1 0 0 1 0 0]]: causal attention between 4 blocks. Tokens of a
+          block can attend all previous blocks and all tokens on the same block.
+
+    Args:
+      input_mask: bool[B, N] true if its part of the input, false if padding.
+      mask_ar: int32[B, N] mask that's 1 where previous tokens cannot depend on
+        it and 0 where it shares the same attention mask as the previous token.
+    """
+    if att_masks.ndim != 2:
+        raise ValueError(att_masks.ndim)
+    if pad_masks.ndim != 2:
+        raise ValueError(pad_masks.ndim)
+
+    cumsum = torch.cumsum(att_masks, dim=1)
+    att_2d_masks = cumsum[:, None, :] <= cumsum[:, :, None]
+    pad_2d_masks = pad_masks[:, None, :] * pad_masks[:, :, None]
+    return att_2d_masks & pad_2d_masks
+
+
+class PI0Pytorch(nn.Module):
+    def __init__(self, config):
+        super().__init__()
+        self.config = config
+        self.pi05 = config.pi05
+
+        paligemma_config = _gemma.get_config(config.paligemma_variant)
+        action_expert_config = _gemma.get_config(config.action_expert_variant)
+
+        self.paligemma_with_expert = PaliGemmaWithExpertModel(
+            paligemma_config,
+            action_expert_config,
+            use_adarms=[False, True] if self.pi05 else [False, False],
+            precision=config.dtype,
+        )
+
+        self.action_in_proj = nn.Linear(config.action_dim, action_expert_config.width)
+        self.action_out_proj = nn.Linear(action_expert_config.width, config.action_dim)
+
+        if self.pi05:
+            self.time_mlp_in = nn.Linear(action_expert_config.width, action_expert_config.width)
+            self.time_mlp_out = nn.Linear(action_expert_config.width, action_expert_config.width)
+        else:
+            self.state_proj = nn.Linear(config.action_dim, action_expert_config.width)
+            self.action_time_mlp_in = nn.Linear(2 * action_expert_config.width, action_expert_config.width)
+            self.action_time_mlp_out = nn.Linear(action_expert_config.width, action_expert_config.width)
+
+        torch.set_float32_matmul_precision("high")
+        if config.pytorch_compile_mode is not None:
+            self.sample_actions = torch.compile(self.sample_actions, mode=config.pytorch_compile_mode)
+
+        # Initialize gradient checkpointing flag
+        self.gradient_checkpointing_enabled = False
+
+        msg = "transformers_replace is not installed correctly. Please install it with `uv pip install transformers==4.53.2` and `cp -r ./src/openpi/models_pytorch/transformers_replace/* .venv/lib/python3.11/site-packages/transformers/`."
+        try:
+            from transformers.models.siglip import check
+
+            if not check.check_whether_transformers_replace_is_installed_correctly():
+                raise ValueError(msg)
+        except ImportError:
+            raise ValueError(msg) from None
+
+    def gradient_checkpointing_enable(self):
+        """Enable gradient checkpointing for memory optimization."""
+        self.gradient_checkpointing_enabled = True
+        self.paligemma_with_expert.paligemma.language_model.gradient_checkpointing = True
+        self.paligemma_with_expert.paligemma.vision_tower.gradient_checkpointing = True
+        self.paligemma_with_expert.gemma_expert.model.gradient_checkpointing = True
+
+        logging.info("Enabled gradient checkpointing for PI0Pytorch model")
+
+    def gradient_checkpointing_disable(self):
+        """Disable gradient checkpointing."""
+        self.gradient_checkpointing_enabled = False
+        self.paligemma_with_expert.paligemma.language_model.gradient_checkpointing = False
+        self.paligemma_with_expert.paligemma.vision_tower.gradient_checkpointing = False
+        self.paligemma_with_expert.gemma_expert.model.gradient_checkpointing = False
+
+        logging.info("Disabled gradient checkpointing for PI0Pytorch model")
+
+    def is_gradient_checkpointing_enabled(self):
+        """Check if gradient checkpointing is enabled."""
+        return self.gradient_checkpointing_enabled
+
+    def _apply_checkpoint(self, func, *args, **kwargs):
+        """Helper method to apply gradient checkpointing if enabled."""
+        if self.gradient_checkpointing_enabled and self.training:
+            return torch.utils.checkpoint.checkpoint(
+                func, *args, use_reentrant=False, preserve_rng_state=False, **kwargs
+            )
+        return func(*args, **kwargs)
+
+    def _prepare_attention_masks_4d(self, att_2d_masks):
+        """Helper method to prepare 4D attention masks for transformer."""
+        att_2d_masks_4d = att_2d_masks[:, None, :, :]
+        return torch.where(att_2d_masks_4d, 0.0, -2.3819763e38)
+
+    def _preprocess_observation(self, observation, *, train=True):
+        """Helper method to preprocess observation."""
+        observation = _preprocessing.preprocess_observation_pytorch(observation, train=train)
+        return (
+            list(observation.images.values()),
+            list(observation.image_masks.values()),
+            observation.tokenized_prompt,
+            observation.tokenized_prompt_mask,
+            observation.state,
+        )
+
+    def sample_noise(self, shape, device):
+        return torch.normal(
+            mean=0.0,
+            std=1.0,
+            size=shape,
+            dtype=torch.float32,
+            device=device,
+        )
+
+    def sample_time(self, bsize, device):
+        time_beta = sample_beta(1.5, 1.0, bsize, device)
+        time = time_beta * 0.999 + 0.001
+        return time.to(dtype=torch.float32, device=device)
+
+    def embed_prefix(
+        self, images, img_masks, lang_tokens, lang_masks
+    ) -> tuple[torch.Tensor, torch.Tensor, torch.Tensor]:
+        """Embed images with SigLIP and language tokens with embedding layer to prepare
+        for PaliGemma transformer processing.
+        """
+        embs = []
+        pad_masks = []
+        att_masks = []
+
+        # Process images
+        for img, img_mask in zip(images, img_masks, strict=True):
+
+            def image_embed_func(img):
+                return self.paligemma_with_expert.embed_image(img)
+
+            img_emb = self._apply_checkpoint(image_embed_func, img)
+
+            bsize, num_img_embs = img_emb.shape[:2]
+
+            embs.append(img_emb)
+            pad_masks.append(img_mask[:, None].expand(bsize, num_img_embs))
+
+            # Create attention masks so that image tokens attend to each other
+            att_masks += [0] * num_img_embs
+
+        # Process language tokens
+        def lang_embed_func(lang_tokens):
+            lang_emb = self.paligemma_with_expert.embed_language_tokens(lang_tokens)
+            lang_emb_dim = lang_emb.shape[-1]
+            return lang_emb * math.sqrt(lang_emb_dim)
+
+        lang_emb = self._apply_checkpoint(lang_embed_func, lang_tokens)
+
+        embs.append(lang_emb)
+        pad_masks.append(lang_masks)
+
+        # full attention between image and language inputs
+        num_lang_embs = lang_emb.shape[1]
+        att_masks += [0] * num_lang_embs
+
+        embs = torch.cat(embs, dim=1)
+        pad_masks = torch.cat(pad_masks, dim=1)
+        att_masks = torch.tensor(att_masks, dtype=torch.bool, device=pad_masks.device)
+
+        # Get batch size from the first dimension of the concatenated tensors
+        bsize = pad_masks.shape[0]
+        att_masks = att_masks[None, :].expand(bsize, len(att_masks))
+
+        return embs, pad_masks, att_masks
+
+    def embed_suffix(self, state, noisy_actions, timestep):
+        """Embed state, noisy_actions, timestep to prepare for Expert Gemma processing."""
+        embs = []
+        pad_masks = []
+        att_masks = []
+
+        if not self.pi05:
+            if self.state_proj.weight.dtype == torch.float32:
+                state = state.to(torch.float32)
+
+            # Embed state
+            def state_proj_func(state):
+                return self.state_proj(state)
+
+            state_emb = self._apply_checkpoint(state_proj_func, state)
+
+            embs.append(state_emb[:, None, :])
+            bsize = state_emb.shape[0]
+            device = state_emb.device
+
+            state_mask = torch.ones(bsize, 1, dtype=torch.bool, device=device)
+            pad_masks.append(state_mask)
+
+            # Set attention masks so that image and language inputs do not attend to state or actions
+            att_masks += [1]
+
+        # Embed timestep using sine-cosine positional encoding with sensitivity in the range [0, 1]
+        time_emb = create_sinusoidal_pos_embedding(
+            timestep, self.action_in_proj.out_features, min_period=4e-3, max_period=4.0, device=timestep.device
+        )
+        time_emb = time_emb.type(dtype=timestep.dtype)
+
+        # Fuse timestep + action information using an MLP
+        def action_proj_func(noisy_actions):
+            return self.action_in_proj(noisy_actions)
+
+        action_emb = self._apply_checkpoint(action_proj_func, noisy_actions)
+
+        if not self.pi05:
+            time_emb = time_emb[:, None, :].expand_as(action_emb)
+            action_time_emb = torch.cat([action_emb, time_emb], dim=2)
+
+            # Apply MLP layers
+            def mlp_func(action_time_emb):
+                x = self.action_time_mlp_in(action_time_emb)
+                x = F.silu(x)  # swish == silu
+                return self.action_time_mlp_out(x)
+
+            action_time_emb = self._apply_checkpoint(mlp_func, action_time_emb)
+            adarms_cond = None
+        else:
+            # time MLP (for adaRMS)
+            def time_mlp_func(time_emb):
+                x = self.time_mlp_in(time_emb)
+                x = F.silu(x)  # swish == silu
+                x = self.time_mlp_out(x)
+                return F.silu(x)
+
+            time_emb = self._apply_checkpoint(time_mlp_func, time_emb)
+            action_time_emb = action_emb
+            adarms_cond = time_emb
+
+        # Add to input tokens
+        embs.append(action_time_emb)
+
+        bsize, action_time_dim = action_time_emb.shape[:2]
+        action_time_mask = torch.ones(bsize, action_time_dim, dtype=torch.bool, device=timestep.device)
+        pad_masks.append(action_time_mask)
+
+        # Set attention masks so that image, language and state inputs do not attend to action tokens
+        att_masks += [1] + ([0] * (self.config.action_horizon - 1))
+
+        embs = torch.cat(embs, dim=1)
+        pad_masks = torch.cat(pad_masks, dim=1)
+        att_masks = torch.tensor(att_masks, dtype=embs.dtype, device=embs.device)
+        att_masks = att_masks[None, :].expand(bsize, len(att_masks))
+
+        return embs, pad_masks, att_masks, adarms_cond
+
+    def forward(self, observation, actions, noise=None, time=None) -> Tensor:
+        """Do a full training forward pass and compute the loss (batch_size x num_steps x num_motors)"""
+        images, img_masks, lang_tokens, lang_masks, state = self._preprocess_observation(observation, train=True)
+
+        if noise is None:
+            noise = self.sample_noise(actions.shape, actions.device)
+
+        if time is None:
+            time = self.sample_time(actions.shape[0], actions.device)
+
+        time_expanded = time[:, None, None]
+        x_t = time_expanded * noise + (1 - time_expanded) * actions
+        u_t = noise - actions
+
+        prefix_embs, prefix_pad_masks, prefix_att_masks = self.embed_prefix(images, img_masks, lang_tokens, lang_masks)
+        suffix_embs, suffix_pad_masks, suffix_att_masks, adarms_cond = self.embed_suffix(state, x_t, time)
+        if (
+            self.paligemma_with_expert.paligemma.language_model.layers[0].self_attn.q_proj.weight.dtype
+            == torch.bfloat16
+        ):
+            suffix_embs = suffix_embs.to(dtype=torch.bfloat16)
+            prefix_embs = prefix_embs.to(dtype=torch.bfloat16)
+
+        pad_masks = torch.cat([prefix_pad_masks, suffix_pad_masks], dim=1)
+        att_masks = torch.cat([prefix_att_masks, suffix_att_masks], dim=1)
+
+        att_2d_masks = make_att_2d_masks(pad_masks, att_masks)
+        position_ids = torch.cumsum(pad_masks, dim=1) - 1
+
+        # Prepare attention masks
+        att_2d_masks_4d = self._prepare_attention_masks_4d(att_2d_masks)
+
+        # Apply gradient checkpointing if enabled
+        def forward_func(prefix_embs, suffix_embs, att_2d_masks_4d, position_ids, adarms_cond):
+            (_, suffix_out), _ = self.paligemma_with_expert.forward(
+                attention_mask=att_2d_masks_4d,
+                position_ids=position_ids,
+                past_key_values=None,
+                inputs_embeds=[prefix_embs, suffix_embs],
+                use_cache=False,
+                adarms_cond=[None, adarms_cond],
+            )
+            return suffix_out
+
+        suffix_out = self._apply_checkpoint(
+            forward_func, prefix_embs, suffix_embs, att_2d_masks_4d, position_ids, adarms_cond
+        )
+
+        suffix_out = suffix_out[:, -self.config.action_horizon :]
+        suffix_out = suffix_out.to(dtype=torch.float32)
+
+        # Apply gradient checkpointing to final action projection if enabled
+        def action_out_proj_func(suffix_out):
+            return self.action_out_proj(suffix_out)
+
+        v_t = self._apply_checkpoint(action_out_proj_func, suffix_out)
+
+        return F.mse_loss(u_t, v_t, reduction="none")
+
+    @torch.no_grad()
+    def sample_actions(self, device, observation, noise=None, num_steps=10) -> Tensor:
+        """Do a full inference forward and compute the action (batch_size x num_steps x num_motors)"""
+        bsize = observation.state.shape[0]
+        if noise is None:
+            actions_shape = (bsize, self.config.action_horizon, self.config.action_dim)
+            noise = self.sample_noise(actions_shape, device)
+
+        images, img_masks, lang_tokens, lang_masks, state = self._preprocess_observation(observation, train=False)
+
+        prefix_embs, prefix_pad_masks, prefix_att_masks = self.embed_prefix(images, img_masks, lang_tokens, lang_masks)
+        prefix_att_2d_masks = make_att_2d_masks(prefix_pad_masks, prefix_att_masks)
+        prefix_position_ids = torch.cumsum(prefix_pad_masks, dim=1) - 1
+
+        # Compute image and language key value cache
+        prefix_att_2d_masks_4d = self._prepare_attention_masks_4d(prefix_att_2d_masks)
+        self.paligemma_with_expert.paligemma.language_model.config._attn_implementation = "eager"  # noqa: SLF001
+
+        _, past_key_values = self.paligemma_with_expert.forward(
+            attention_mask=prefix_att_2d_masks_4d,
+            position_ids=prefix_position_ids,
+            past_key_values=None,
+            inputs_embeds=[prefix_embs, None],
+            use_cache=True,
+        )
+
+        dt = -1.0 / num_steps
+        dt = torch.tensor(dt, dtype=torch.float32, device=device)
+
+        x_t = noise
+        time = torch.tensor(1.0, dtype=torch.float32, device=device)
+        while time >= -dt / 2:
+            expanded_time = time.expand(bsize)
+            v_t = self.denoise_step(
+                state,
+                prefix_pad_masks,
+                past_key_values,
+                x_t,
+                expanded_time,
+            )
+
+            # Euler step - use new tensor assignment instead of in-place operation
+            x_t = x_t + dt * v_t
+            time += dt
+        return x_t
+
+    def denoise_step(
+        self,
+        state,
+        prefix_pad_masks,
+        past_key_values,
+        x_t,
+        timestep,
+    ):
+        """Apply one denoising step of the noise `x_t` at a given timestep."""
+        suffix_embs, suffix_pad_masks, suffix_att_masks, adarms_cond = self.embed_suffix(state, x_t, timestep)
+
+        suffix_len = suffix_pad_masks.shape[1]
+        batch_size = prefix_pad_masks.shape[0]
+        prefix_len = prefix_pad_masks.shape[1]
+
+        prefix_pad_2d_masks = prefix_pad_masks[:, None, :].expand(batch_size, suffix_len, prefix_len)
+
+        suffix_att_2d_masks = make_att_2d_masks(suffix_pad_masks, suffix_att_masks)
+
+        full_att_2d_masks = torch.cat([prefix_pad_2d_masks, suffix_att_2d_masks], dim=2)
+
+        prefix_offsets = torch.sum(prefix_pad_masks, dim=-1)[:, None]
+        position_ids = prefix_offsets + torch.cumsum(suffix_pad_masks, dim=1) - 1
+
+        # Prepare attention masks
+        full_att_2d_masks_4d = self._prepare_attention_masks_4d(full_att_2d_masks)
+        self.paligemma_with_expert.gemma_expert.model.config._attn_implementation = "eager"  # noqa: SLF001
+
+        outputs_embeds, _ = self.paligemma_with_expert.forward(
+            attention_mask=full_att_2d_masks_4d,
+            position_ids=position_ids,
+            past_key_values=past_key_values,
+            inputs_embeds=[None, suffix_embs],
+            use_cache=False,
+            adarms_cond=[None, adarms_cond],
+        )
+
+        suffix_out = outputs_embeds[1]
+        suffix_out = suffix_out[:, -self.config.action_horizon :]
+        suffix_out = suffix_out.to(dtype=torch.float32)
+        return self.action_out_proj(suffix_out)

real_inference/source/XPolicyLab/policy/Pi_05/openpi/src/openpi/models_pytorch/transformers_replace/models/gemma/configuration_gemma.py

@@ -0,0 +1,173 @@
+#                🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨
+#           This file was automatically generated from src/transformers/models/gemma/modular_gemma.py.
+#               Do NOT edit this file manually as any edits will be overwritten by the generation of
+#             the file from the modular. If any change should be done, please apply the change to the
+#                          modular_gemma.py file directly. One of our CI enforces this.
+#                🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨
+# coding=utf-8
+# Copyright 2024 Google Inc. HuggingFace Inc. team. All rights reserved.
+#
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+from typing import Optional
+from ...configuration_utils import PretrainedConfig
+
+
+class GemmaConfig(PretrainedConfig):
+    r"""
+    This is the configuration class to store the configuration of a [`GemmaModel`]. It is used to instantiate an Gemma
+    model according to the specified arguments, defining the model architecture. Instantiating a configuration with the
+    defaults will yield a similar configuration to that of the Gemma-7B.
+    e.g. [google/gemma-7b](https://huggingface.co/google/gemma-7b)
+    Configuration objects inherit from [`PretrainedConfig`] and can be used to control the model outputs. Read the
+    documentation from [`PretrainedConfig`] for more information.
+    Args:
+        vocab_size (`int`, *optional*, defaults to 256000):
+            Vocabulary size of the Gemma model. Defines the number of different tokens that can be represented by the
+            `inputs_ids` passed when calling [`GemmaModel`]
+        hidden_size (`int`, *optional*, defaults to 3072):
+            Dimension of the hidden representations.
+        intermediate_size (`int`, *optional*, defaults to 24576):
+            Dimension of the MLP representations.
+        num_hidden_layers (`int`, *optional*, defaults to 28):
+            Number of hidden layers in the Transformer decoder.
+        num_attention_heads (`int`, *optional*, defaults to 16):
+            Number of attention heads for each attention layer in the Transformer decoder.
+        num_key_value_heads (`int`, *optional*, defaults to 16):
+            This is the number of key_value heads that should be used to implement Grouped Query Attention. If
+            `num_key_value_heads=num_attention_heads`, the model will use Multi Head Attention (MHA), if
+            `num_key_value_heads=1` the model will use Multi Query Attention (MQA) otherwise GQA is used. When
+            converting a multi-head checkpoint to a GQA checkpoint, each group key and value head should be constructed
+            by meanpooling all the original heads within that group. For more details, check out [this
+            paper](https://huggingface.co/papers/2305.13245). If it is not specified, will default to
+            `num_attention_heads`.
+        head_dim (`int`, *optional*, defaults to 256):
+            The attention head dimension.
+        hidden_act (`str` or `function`, *optional*, defaults to `"gelu_pytorch_tanh"`):
+            The legacy activation function. It is overwritten by the `hidden_activation`.
+        hidden_activation (`str` or `function`, *optional*):
+            The non-linear activation function (function or string) in the decoder. Will default to `"gelu_pytorch_tanh"`
+            if not specified. `"gelu_pytorch_tanh"` uses an approximation of the `"gelu"` activation function.
+        max_position_embeddings (`int`, *optional*, defaults to 8192):
+            The maximum sequence length that this model might ever be used with.
+        initializer_range (`float`, *optional*, defaults to 0.02):
+            The standard deviation of the truncated_normal_initializer for initializing all weight matrices.
+        rms_norm_eps (`float`, *optional*, defaults to 1e-06):
+            The epsilon used by the rms normalization layers.
+        use_cache (`bool`, *optional*, defaults to `True`):
+            Whether or not the model should return the last key/values attentions (not used by all models). Only
+            relevant if `config.is_decoder=True`.
+        pad_token_id (`int`, *optional*, defaults to 0):
+            Padding token id.
+        eos_token_id (`int`, *optional*, defaults to 1):
+            End of stream token id.
+        bos_token_id (`int`, *optional*, defaults to 2):
+            Beginning of stream token id.
+        tie_word_embeddings (`bool`, *optional*, defaults to `True`):
+            Whether to tie weight embeddings
+        rope_theta (`float`, *optional*, defaults to 10000.0):
+            The base period of the RoPE embeddings.
+        attention_bias (`bool`, defaults to `False`, *optional*, defaults to `False`):
+            Whether to use a bias in the query, key, value and output projection layers during self-attention.
+        attention_dropout (`float`, *optional*, defaults to 0.0):
+            The dropout ratio for the attention probabilities.
+        use_adarms (`bool`, *optional*, defaults to `False`):
+            Whether to use ADARMS.
+        adarms_cond_dim (`int`, *optional*, defaults to `None`):
+            The dimension of the ADARMS condition.
+    ```python
+    >>> from transformers import GemmaModel, GemmaConfig
+    >>> # Initializing a Gemma gemma-7b style configuration
+    >>> configuration = GemmaConfig()
+    >>> # Initializing a model from the gemma-7b style configuration
+    >>> model = GemmaModel(configuration)
+    >>> # Accessing the model configuration
+    >>> configuration = model.config
+    ```"""
+
+    model_type = "gemma"
+    keys_to_ignore_at_inference = ["past_key_values"]
+    base_model_tp_plan = {
+        "layers.*.self_attn.q_proj": "colwise",
+        "layers.*.self_attn.k_proj": "colwise",
+        "layers.*.self_attn.v_proj": "colwise",
+        "layers.*.self_attn.o_proj": "rowwise",
+        "layers.*.mlp.gate_proj": "colwise",
+        "layers.*.mlp.up_proj": "colwise",
+        "layers.*.mlp.down_proj": "rowwise",
+    }
+    base_model_pp_plan = {
+        "embed_tokens": (["input_ids"], ["inputs_embeds"]),
+        "layers": (["hidden_states", "attention_mask"], ["hidden_states"]),
+        "norm": (["hidden_states"], ["hidden_states"]),
+    }
+
+    def __init__(
+        self,
+        vocab_size=256000,
+        hidden_size=3072,
+        intermediate_size=24576,
+        num_hidden_layers=28,
+        num_attention_heads=16,
+        num_key_value_heads=16,
+        head_dim=256,
+        hidden_act="gelu_pytorch_tanh",
+        hidden_activation=None,
+        max_position_embeddings=8192,
+        initializer_range=0.02,
+        rms_norm_eps=1e-6,
+        use_cache=True,
+        pad_token_id=0,
+        eos_token_id=1,
+        bos_token_id=2,
+        tie_word_embeddings=True,
+        rope_theta=10000.0,
+        attention_bias=False,
+        attention_dropout=0.0,
+        use_adarms: bool = False,
+        adarms_cond_dim: Optional[int] = None,
+        **kwargs,
+    ):
+        self.vocab_size = vocab_size
+        self.max_position_embeddings = max_position_embeddings
+        self.hidden_size = hidden_size
+        self.intermediate_size = intermediate_size
+        self.num_hidden_layers = num_hidden_layers
+        self.num_attention_heads = num_attention_heads
+        self.head_dim = head_dim
+        self.num_key_value_heads = num_key_value_heads
+        self.hidden_act = hidden_act
+        self.hidden_activation = hidden_activation
+        self.initializer_range = initializer_range
+        self.rms_norm_eps = rms_norm_eps
+        self.use_cache = use_cache
+        self.rope_theta = rope_theta
+        self.attention_bias = attention_bias
+        self.attention_dropout = attention_dropout
+        self.use_adarms = use_adarms
+        self.adarms_cond_dim = adarms_cond_dim
+
+        # Set default for adarms_cond_dim if use_adarms is True
+        if self.use_adarms and self.adarms_cond_dim is None:
+            self.adarms_cond_dim = self.hidden_size
+
+        super().__init__(
+            pad_token_id=pad_token_id,
+            bos_token_id=bos_token_id,
+            eos_token_id=eos_token_id,
+            tie_word_embeddings=tie_word_embeddings,
+            **kwargs,
+        )
+
+
+__all__ = ["GemmaConfig"]

real_inference/source/XPolicyLab/policy/Pi_05/openpi/src/openpi/models_pytorch/transformers_replace/models/gemma/modeling_gemma.py

@@ -0,0 +1,862 @@
+#                🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨
+#           This file was automatically generated from src/transformers/models/gemma/modular_gemma.py.
+#               Do NOT edit this file manually as any edits will be overwritten by the generation of
+#             the file from the modular. If any change should be done, please apply the change to the
+#                          modular_gemma.py file directly. One of our CI enforces this.
+#                🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨
+# coding=utf-8
+# Copyright 2024 Google Inc. HuggingFace Inc. team. All rights reserved.
+#
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+from typing import Callable, Optional, Union
+
+import torch
+from torch import nn
+
+from ...activations import ACT2FN
+from ...cache_utils import Cache, DynamicCache
+from ...generation import GenerationMixin
+from ...masking_utils import create_causal_mask
+from ...modeling_flash_attention_utils import FlashAttentionKwargs
+from ...modeling_layers import GradientCheckpointingLayer
+from ...modeling_outputs import (
+    BaseModelOutputWithPast,
+    CausalLMOutputWithPast,
+    SequenceClassifierOutputWithPast,
+    TokenClassifierOutput,
+)
+from ...modeling_rope_utils import ROPE_INIT_FUNCTIONS, dynamic_rope_update
+from ...modeling_utils import ALL_ATTENTION_FUNCTIONS, PreTrainedModel
+from ...processing_utils import Unpack
+from ...utils import LossKwargs, auto_docstring, can_return_tuple, logging
+from .configuration_gemma import GemmaConfig
+
+
+logger = logging.get_logger(__name__)
+
+
+class GemmaRMSNorm(nn.Module):
+    def __init__(self, dim: int, eps: float = 1e-6, cond_dim: Optional[int] = None):
+        super().__init__()
+        self.eps = eps
+        self.dim = dim
+        self.cond_dim = cond_dim
+        
+        # Dense layer for adaptive normalization (if cond_dim is provided)
+        if cond_dim is not None:
+            #self.dense = nn.Linear(cond_dim, dim * 3, bias=True, dtype=torch.bfloat16)
+            self.dense = nn.Linear(cond_dim, dim * 3, bias=True)
+            # Initialize with zeros (matches source implementation)
+            nn.init.zeros_(self.dense.weight)
+        else:
+            self.weight = nn.Parameter(torch.zeros(dim, dtype=torch.bfloat16))
+            self.dense = None
+
+    def _norm(self, x):
+        # Compute variance in float32 (like the source implementation)
+        var = torch.mean(torch.square(x.float()), dim=-1, keepdim=True)
+        # Compute normalization in float32
+        normed_inputs = x * torch.rsqrt(var + self.eps)
+        return normed_inputs
+
+    def forward(self, x, cond=None):
+        dtype = x.dtype  # original dtype, could be half-precision
+        normed_inputs = self._norm(x)
+        
+        if cond is None or self.dense is None:
+            # regular RMSNorm
+            # scale by learned parameter in float32 (matches source implementation)
+            normed_inputs = normed_inputs * (1.0 + self.weight.float())
+            return normed_inputs.to(dtype), None  # return in original dtype with None gate
+        
+        # adaptive RMSNorm (if cond is provided and dense layer exists)
+        if cond.shape[-1] != self.cond_dim:
+            raise ValueError(f"Expected cond dimension {self.cond_dim}, got {cond.shape[-1]}")
+        
+        #self.dense.to(dtype=torch.bfloat16).to(dtype=torch.float32)
+        modulation = self.dense(cond)
+        # Reshape modulation to broadcast properly: [batch, 1, features] for [batch, seq, features]
+        if len(x.shape) == 3:  # [batch, seq, features]
+            modulation = modulation.unsqueeze(1)
+        
+        scale, shift, gate = torch.chunk(modulation, 3, dim=-1)
+        
+        # Apply adaptive normalization: use model weight dtype to ensure compatibility
+        # model_dtype = self.dense.weight.dtype  # Use the model's dtype (bfloat16)
+        # scale = scale.to(model_dtype)
+        # shift = shift.to(model_dtype)
+        # gate = gate.to(model_dtype)
+        # normed_inputs = normed_inputs.to(model_dtype)  # Convert normed_inputs to model dtype
+        
+        normed_inputs = normed_inputs * (1 + scale.to(torch.float32)) + shift.to(torch.float32)
+
+        return normed_inputs.to(dtype), gate.to(dtype)
+
+    def extra_repr(self):
+        repr_str = f"{tuple(self.weight.shape)}, eps={self.eps}"
+        if self.dense is not None:
+            repr_str += f", adaptive=True, cond_dim={self.cond_dim}"
+        return repr_str
+
+
+class GemmaMLP(nn.Module):
+    def __init__(self, config):
+        super().__init__()
+        self.config = config
+        self.hidden_size = config.hidden_size
+        self.intermediate_size = config.intermediate_size
+        self.gate_proj = nn.Linear(self.hidden_size, self.intermediate_size, bias=False)
+        self.up_proj = nn.Linear(self.hidden_size, self.intermediate_size, bias=False)
+        self.down_proj = nn.Linear(self.intermediate_size, self.hidden_size, bias=False)
+        self.act_fn = ACT2FN[config.hidden_act]
+
+    def forward(self, x):
+        down_proj = self.down_proj(self.act_fn(self.gate_proj(x)) * self.up_proj(x))
+        return down_proj
+
+
+class GemmaRotaryEmbedding(nn.Module):
+    def __init__(self, config: GemmaConfig, device=None):
+        super().__init__()
+        # BC: "rope_type" was originally "type"
+        if hasattr(config, "rope_scaling") and config.rope_scaling is not None:
+            self.rope_type = config.rope_scaling.get("rope_type", config.rope_scaling.get("type"))
+        else:
+            self.rope_type = "default"
+        self.max_seq_len_cached = config.max_position_embeddings
+        self.original_max_seq_len = config.max_position_embeddings
+
+        self.config = config
+        self.rope_init_fn = ROPE_INIT_FUNCTIONS[self.rope_type]
+
+        inv_freq, self.attention_scaling = self.rope_init_fn(self.config, device)
+        self.register_buffer("inv_freq", inv_freq, persistent=False)
+        self.original_inv_freq = self.inv_freq
+
+    @torch.no_grad()
+    @dynamic_rope_update  # power user: used with advanced RoPE types (e.g. dynamic rope)
+    def forward(self, x, position_ids):
+        inv_freq_expanded = self.inv_freq[None, :, None].float().expand(position_ids.shape[0], -1, 1).to(x.device)
+        position_ids_expanded = position_ids[:, None, :].float()
+
+        device_type = x.device.type if isinstance(x.device.type, str) and x.device.type != "mps" else "cpu"
+        with torch.autocast(device_type=device_type, enabled=False):  # Force float32
+            freqs = (inv_freq_expanded.float() @ position_ids_expanded.float()).transpose(1, 2)
+            emb = torch.cat((freqs, freqs), dim=-1)
+            cos = emb.cos() * self.attention_scaling
+            sin = emb.sin() * self.attention_scaling
+
+        return cos.to(dtype=x.dtype), sin.to(dtype=x.dtype)
+
+
+def rotate_half(x):
+    """Rotates half the hidden dims of the input."""
+    x1 = x[..., : x.shape[-1] // 2]
+    x2 = x[..., x.shape[-1] // 2 :]
+    return torch.cat((-x2, x1), dim=-1)
+
+
+def apply_rotary_pos_emb(q, k, cos, sin, position_ids=None, unsqueeze_dim=1):
+    """Applies Rotary Position Embedding to the query and key tensors.
+
+    Args:
+        q (`torch.Tensor`): The query tensor.
+        k (`torch.Tensor`): The key tensor.
+        cos (`torch.Tensor`): The cosine part of the rotary embedding.
+        sin (`torch.Tensor`): The sine part of the rotary embedding.
+        position_ids (`torch.Tensor`, *optional*):
+            Deprecated and unused.
+        unsqueeze_dim (`int`, *optional*, defaults to 1):
+            The 'unsqueeze_dim' argument specifies the dimension along which to unsqueeze cos[position_ids] and
+            sin[position_ids] so that they can be properly broadcasted to the dimensions of q and k. For example, note
+            that cos[position_ids] and sin[position_ids] have the shape [batch_size, seq_len, head_dim]. Then, if q and
+            k have the shape [batch_size, heads, seq_len, head_dim], then setting unsqueeze_dim=1 makes
+            cos[position_ids] and sin[position_ids] broadcastable to the shapes of q and k. Similarly, if q and k have
+            the shape [batch_size, seq_len, heads, head_dim], then set unsqueeze_dim=2.
+    Returns:
+        `tuple(torch.Tensor)` comprising of the query and key tensors rotated using the Rotary Position Embedding.
+    """
+    cos = cos.unsqueeze(unsqueeze_dim)
+    sin = sin.unsqueeze(unsqueeze_dim)
+    q_embed = (q * cos) + (rotate_half(q) * sin)
+    k_embed = (k * cos) + (rotate_half(k) * sin)
+    return q_embed, k_embed
+
+
+def repeat_kv(hidden_states: torch.Tensor, n_rep: int) -> torch.Tensor:
+    """
+    This is the equivalent of torch.repeat_interleave(x, dim=1, repeats=n_rep). The hidden states go from (batch,
+    num_key_value_heads, seqlen, head_dim) to (batch, num_attention_heads, seqlen, head_dim)
+    """
+    batch, num_key_value_heads, slen, head_dim = hidden_states.shape
+    if n_rep == 1:
+        return hidden_states
+    hidden_states = hidden_states[:, :, None, :, :].expand(batch, num_key_value_heads, n_rep, slen, head_dim)
+    return hidden_states.reshape(batch, num_key_value_heads * n_rep, slen, head_dim)
+
+
+def _gated_residual(x, y, gate):
+    """
+    Applies gated residual connection with optional gate parameter.
+    
+    Args:
+        x: Input tensor (residual)
+        y: Output tensor to be added
+        gate: Optional gate tensor to modulate the addition
+        
+    Returns:
+        x + y if gate is None, otherwise x + y * gate
+    """
+    if x is None and y is None:
+        return None
+    if x is None or y is None:
+        return x if x is not None else y
+    if gate is None:
+        return x + y
+    return x + y * gate
+
+
+def eager_attention_forward(
+    module: nn.Module,
+    query: torch.Tensor,
+    key: torch.Tensor,
+    value: torch.Tensor,
+    attention_mask: Optional[torch.Tensor],
+    scaling: float,
+    dropout: float = 0.0,
+    **kwargs,
+):
+    key_states = repeat_kv(key, module.num_key_value_groups)
+    value_states = repeat_kv(value, module.num_key_value_groups)
+
+    attn_weights = torch.matmul(query, key_states.transpose(2, 3)) * scaling
+    if attention_mask is not None:
+        causal_mask = attention_mask[:, :, :, : key_states.shape[-2]]
+        attn_weights = attn_weights + causal_mask
+
+    attn_weights = nn.functional.softmax(attn_weights, dim=-1, dtype=torch.float32).to(query.dtype)
+    attn_weights = nn.functional.dropout(attn_weights, p=dropout, training=module.training)
+    attn_output = torch.matmul(attn_weights, value_states)
+    attn_output = attn_output.transpose(1, 2).contiguous()
+
+    return attn_output, attn_weights
+
+
+class GemmaAttention(nn.Module):
+    """Multi-headed attention from 'Attention Is All You Need' paper"""
+
+    def __init__(self, config: GemmaConfig, layer_idx: int):
+        super().__init__()
+        self.config = config
+        self.layer_idx = layer_idx
+        self.head_dim = getattr(config, "head_dim", config.hidden_size // config.num_attention_heads)
+        self.num_key_value_groups = config.num_attention_heads // config.num_key_value_heads
+        self.scaling = self.head_dim**-0.5
+        self.attention_dropout = config.attention_dropout
+        self.is_causal = True
+
+        self.q_proj = nn.Linear(
+            config.hidden_size, config.num_attention_heads * self.head_dim, bias=config.attention_bias
+        )
+        self.k_proj = nn.Linear(
+            config.hidden_size, config.num_key_value_heads * self.head_dim, bias=config.attention_bias
+        )
+        self.v_proj = nn.Linear(
+            config.hidden_size, config.num_key_value_heads * self.head_dim, bias=config.attention_bias
+        )
+        self.o_proj = nn.Linear(
+            config.num_attention_heads * self.head_dim, config.hidden_size, bias=config.attention_bias
+        )
+
+    def forward(
+        self,
+        hidden_states: torch.Tensor,
+        position_embeddings: tuple[torch.Tensor, torch.Tensor],
+        attention_mask: Optional[torch.Tensor],
+        past_key_value: Optional[Cache] = None,
+        cache_position: Optional[torch.LongTensor] = None,
+        use_cache: bool = False,
+        **kwargs: Unpack[FlashAttentionKwargs],
+    ) -> tuple[torch.Tensor, Optional[torch.Tensor], Optional[tuple[torch.Tensor]]]:        
+        input_shape = hidden_states.shape[:-1]
+        hidden_shape = (*input_shape, -1, self.head_dim)
+
+        query_states = self.q_proj(hidden_states).view(hidden_shape).transpose(1, 2)
+        key_states = self.k_proj(hidden_states).view(hidden_shape).transpose(1, 2)
+        value_states = self.v_proj(hidden_states).view(hidden_shape).transpose(1, 2)
+
+        cos, sin = position_embeddings
+        query_states, key_states = apply_rotary_pos_emb(query_states, key_states, cos, sin)
+
+        # Use cache if provided
+        if past_key_value is not None:
+            if use_cache:
+                # sin and cos are specific to RoPE models; cache_position needed for the static cache
+                cache_kwargs = {"sin": sin, "cos": cos, "cache_position": cache_position}
+                key_states, value_states = past_key_value.update(key_states, value_states, self.layer_idx, cache_kwargs)
+            else:
+                key_states = torch.cat([past_key_value[self.layer_idx][0], key_states], dim=2)
+                value_states = torch.cat([past_key_value[self.layer_idx][1], value_states], dim=2)
+
+        attention_interface: Callable = eager_attention_forward
+        if self.config._attn_implementation != "eager":
+            attention_interface = ALL_ATTENTION_FUNCTIONS[self.config._attn_implementation]
+
+        attn_output, attn_weights = attention_interface(
+            self,
+            query_states,
+            key_states,
+            value_states,
+            attention_mask,
+            dropout=0.0 if not self.training else self.attention_dropout,
+            scaling=self.scaling,
+            **kwargs,
+        )
+
+        attn_output = attn_output.reshape(*input_shape, -1).contiguous()
+        attn_output = self.o_proj(attn_output)
+        return attn_output, attn_weights
+
+
+class GemmaDecoderLayer(GradientCheckpointingLayer):
+    def __init__(self, config: GemmaConfig, layer_idx: int):
+        super().__init__()
+        self.hidden_size = config.hidden_size
+
+        self.self_attn = GemmaAttention(config=config, layer_idx=layer_idx)
+
+        self.mlp = GemmaMLP(config)
+        cond_dim = getattr(config, 'adarms_cond_dim', None) if getattr(config, 'use_adarms', False) else None
+        self.input_layernorm = GemmaRMSNorm(config.hidden_size, eps=config.rms_norm_eps, cond_dim=cond_dim)
+        self.post_attention_layernorm = GemmaRMSNorm(config.hidden_size, eps=config.rms_norm_eps, cond_dim=cond_dim)
+
+    def forward(
+        self,
+        hidden_states: torch.Tensor,
+        attention_mask: Optional[torch.Tensor] = None,
+        position_ids: Optional[torch.LongTensor] = None,
+        past_key_value: Optional[Cache] = None,
+        output_attentions: Optional[bool] = False,
+        use_cache: Optional[bool] = False,
+        cache_position: Optional[torch.LongTensor] = None,
+        position_embeddings: Optional[tuple[torch.Tensor, torch.Tensor]] = None,  # necessary, but kept here for BC
+        adarms_cond: Optional[torch.Tensor] = None,
+        **kwargs: Unpack[FlashAttentionKwargs],
+    ) -> tuple[torch.FloatTensor, Optional[tuple[torch.FloatTensor, torch.FloatTensor]]]:
+        residual = hidden_states
+        hidden_states, gate = self.input_layernorm(hidden_states, adarms_cond)
+
+        # Self Attention
+        hidden_states, self_attn_weights = self.self_attn(
+            hidden_states=hidden_states,
+            attention_mask=attention_mask,
+            position_ids=position_ids,
+            past_key_value=past_key_value,
+            output_attentions=output_attentions,
+            use_cache=use_cache,
+            cache_position=cache_position,
+            position_embeddings=position_embeddings,
+            **kwargs,
+        )
+        hidden_states = _gated_residual(residual, hidden_states, gate)
+
+        # Fully Connected
+        residual = hidden_states
+        hidden_states, gate = self.post_attention_layernorm(hidden_states, adarms_cond)
+        hidden_states = self.mlp(hidden_states)
+        hidden_states = _gated_residual(residual, hidden_states, gate)
+
+        outputs = (hidden_states,)
+        if output_attentions:
+            outputs += (self_attn_weights,)
+
+        return outputs
+
+
+@auto_docstring
+class GemmaPreTrainedModel(PreTrainedModel):
+    config_class = GemmaConfig
+    base_model_prefix = "model"
+    supports_gradient_checkpointing = True
+    _no_split_modules = ["GemmaDecoderLayer"]
+    _skip_keys_device_placement = ["past_key_values"]
+    _supports_flash_attn_3 = True
+    _supports_flash_attn_2 = True
+    _supports_sdpa = True
+    _supports_flex_attn = True
+    _supports_cache_class = True
+    _supports_quantized_cache = True
+    _supports_static_cache = True
+    _supports_attention_backend = True
+
+    def _init_weights(self, module):
+        std = self.config.initializer_range
+        if isinstance(module, nn.Linear):
+            module.weight.data.normal_(mean=0.0, std=std)
+            if module.bias is not None:
+                module.bias.data.zero_()
+        elif isinstance(module, nn.Embedding):
+            module.weight.data.normal_(mean=0.0, std=std)
+            if module.padding_idx is not None:
+                module.weight.data[module.padding_idx].zero_()
+        elif isinstance(module, GemmaRMSNorm):
+            if hasattr(module, 'weight'):
+                module.weight.data.fill_(1.0)
+
+
+@auto_docstring
+class GemmaModel(GemmaPreTrainedModel):
+    def __init__(self, config: GemmaConfig):
+        super().__init__(config)
+        self.padding_idx = config.pad_token_id
+        self.vocab_size = config.vocab_size
+
+        self.embed_tokens = nn.Embedding(config.vocab_size, config.hidden_size, self.padding_idx)
+        self.layers = nn.ModuleList(
+            [GemmaDecoderLayer(config, layer_idx) for layer_idx in range(config.num_hidden_layers)]
+        )
+
+        cond_dim = getattr(config, 'adarms_cond_dim', None) if getattr(config, 'use_adarms', False) else None
+        self.norm = GemmaRMSNorm(config.hidden_size, eps=config.rms_norm_eps, cond_dim=cond_dim)
+        self.rotary_emb = GemmaRotaryEmbedding(config=config)
+        self.gradient_checkpointing = False
+
+        # Initialize weights and apply final processing
+        self.post_init()
+
+    def get_input_embeddings(self):
+        return self.embed_tokens
+
+    def set_input_embeddings(self, value):
+        self.embed_tokens = value
+
+    @can_return_tuple
+    @auto_docstring
+    def forward(
+        self,
+        input_ids: Optional[torch.LongTensor] = None,
+        attention_mask: Optional[torch.Tensor] = None,
+        position_ids: Optional[torch.LongTensor] = None,
+        past_key_values: Optional[Cache] = None,
+        inputs_embeds: Optional[torch.FloatTensor] = None,
+        use_cache: Optional[bool] = None,
+        output_attentions: Optional[bool] = None,
+        output_hidden_states: Optional[bool] = None,
+        cache_position: Optional[torch.LongTensor] = None,
+        adarms_cond: Optional[torch.Tensor] = None,
+        **kwargs: Unpack[FlashAttentionKwargs],
+    ) -> BaseModelOutputWithPast:
+        """
+        adarms_cond (`torch.Tensor` of shape `(batch_size, cond_dim)`, *optional*):
+            Condition for ADARMS.
+        """
+        output_attentions = output_attentions if output_attentions is not None else self.config.output_attentions
+        output_hidden_states = (
+            output_hidden_states if output_hidden_states is not None else self.config.output_hidden_states
+        )
+        use_cache = use_cache if use_cache is not None else self.config.use_cache
+
+        if (input_ids is None) ^ (inputs_embeds is not None):
+            raise ValueError("You must specify exactly one of input_ids or inputs_embeds")
+
+        if self.gradient_checkpointing and self.training and use_cache:
+            logger.warning_once(
+                "`use_cache=True` is incompatible with gradient checkpointing. Setting `use_cache=False`."
+            )
+            use_cache = False
+
+        if inputs_embeds is None:
+            inputs_embeds = self.embed_tokens(input_ids)
+
+        if use_cache and past_key_values is None:
+            past_key_values = DynamicCache()
+
+        if cache_position is None:
+            past_seen_tokens = past_key_values.get_seq_length() if past_key_values is not None else 0
+            cache_position = torch.arange(
+                past_seen_tokens, past_seen_tokens + inputs_embeds.shape[1], device=inputs_embeds.device
+            )
+
+        if position_ids is None:
+            position_ids = cache_position.unsqueeze(0)
+
+        causal_mask = create_causal_mask(
+            config=self.config,
+            input_embeds=inputs_embeds,
+            attention_mask=attention_mask,
+            cache_position=cache_position,
+            past_key_values=past_key_values,
+            position_ids=position_ids,
+        )
+
+        # embed positions
+        hidden_states = inputs_embeds
+        # Convert to bfloat16 if the first layer uses bfloat16
+        if len(self.layers) > 0 and self.layers[0].self_attn.q_proj.weight.dtype == torch.bfloat16:
+            hidden_states = hidden_states.to(torch.bfloat16)
+
+        # create position embeddings to be shared across the decoder layers
+        position_embeddings = self.rotary_emb(hidden_states, position_ids)
+
+        # normalized
+        # Gemma downcasts the below to float16, causing sqrt(3072)=55.4256 to become 55.5
+        # See https://github.com/huggingface/transformers/pull/29402
+        normalizer = torch.tensor(self.config.hidden_size**0.5, dtype=hidden_states.dtype)
+        #hidden_states = hidden_states * normalizer
+
+        # decoder layers
+        all_hidden_states = () if output_hidden_states else None
+        all_self_attns = () if output_attentions else None
+
+        for decoder_layer in self.layers[: self.config.num_hidden_layers]:
+            if output_hidden_states:
+                all_hidden_states += (hidden_states,)
+
+            layer_outputs = decoder_layer(
+                hidden_states,
+                attention_mask=causal_mask,
+                position_ids=position_ids,
+                past_key_value=past_key_values,
+                output_attentions=output_attentions,
+                use_cache=use_cache,
+                cache_position=cache_position,
+                position_embeddings=position_embeddings,
+                adarms_cond=adarms_cond,
+                **kwargs,
+            )
+
+            hidden_states = layer_outputs[0]
+
+            if output_attentions:
+                all_self_attns += (layer_outputs[1],)
+
+        hidden_states, _ = self.norm(hidden_states, adarms_cond)
+
+        # add hidden states from the last decoder layer
+        if output_hidden_states:
+            all_hidden_states += (hidden_states,)
+
+        return BaseModelOutputWithPast(
+            last_hidden_state=hidden_states,
+            past_key_values=past_key_values if use_cache else None,
+            hidden_states=all_hidden_states,
+            attentions=all_self_attns,
+        )
+
+
+class KwargsForCausalLM(FlashAttentionKwargs, LossKwargs): ...
+
+
+@auto_docstring
+class GemmaForCausalLM(GemmaPreTrainedModel, GenerationMixin):
+    _tied_weights_keys = ["lm_head.weight"]
+    _tp_plan = {"lm_head": "colwise_rep"}
+    _pp_plan = {"lm_head": (["hidden_states"], ["logits"])}
+
+    def __init__(self, config):
+        super().__init__(config)
+        self.model = GemmaModel(config)
+        self.vocab_size = config.vocab_size
+        self.lm_head = nn.Linear(config.hidden_size, config.vocab_size, bias=False)
+
+        # Initialize weights and apply final processing
+        self.post_init()
+
+    def get_input_embeddings(self):
+        return self.model.embed_tokens
+
+    def set_input_embeddings(self, value):
+        self.model.embed_tokens = value
+
+    def get_output_embeddings(self):
+        return self.lm_head
+
+    def set_output_embeddings(self, new_embeddings):
+        self.lm_head = new_embeddings
+
+    def set_decoder(self, decoder):
+        self.model = decoder
+
+    def get_decoder(self):
+        return self.model
+
+    @can_return_tuple
+    @auto_docstring
+    def forward(
+        self,
+        input_ids: Optional[torch.LongTensor] = None,
+        attention_mask: Optional[torch.Tensor] = None,
+        position_ids: Optional[torch.LongTensor] = None,
+        past_key_values: Optional[Cache] = None,
+        inputs_embeds: Optional[torch.FloatTensor] = None,
+        labels: Optional[torch.LongTensor] = None,
+        use_cache: Optional[bool] = None,
+        output_attentions: Optional[bool] = None,
+        output_hidden_states: Optional[bool] = None,
+        cache_position: Optional[torch.LongTensor] = None,
+        logits_to_keep: Union[int, torch.Tensor] = 0,
+        adarms_cond: Optional[torch.Tensor] = None,
+        **kwargs: Unpack[KwargsForCausalLM],
+    ) -> CausalLMOutputWithPast:
+        r"""
+        labels (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*):
+            Labels for computing the masked language modeling loss. Indices should either be in `[0, ...,
+            config.vocab_size]` or -100 (see `input_ids` docstring). Tokens with indices set to `-100` are ignored
+            (masked), the loss is only computed for the tokens with labels in `[0, ..., config.vocab_size]`.
+
+        adarms_cond (`torch.Tensor` of shape `(batch_size, cond_dim)`, *optional*):
+            Condition for ADARMS.
+
+        Example:
+
+        ```python
+        >>> from transformers import AutoTokenizer, GemmaForCausalLM
+
+        >>> model = GemmaForCausalLM.from_pretrained("google/gemma-7b")
+        >>> tokenizer = AutoTokenizer.from_pretrained("google/gemma-7b")
+
+        >>> prompt = "What is your favorite condiment?"
+        >>> inputs = tokenizer(prompt, return_tensors="pt")
+
+        >>> # Generate
+        >>> generate_ids = model.generate(inputs.input_ids, max_length=30)
+        >>> tokenizer.batch_decode(generate_ids, skip_special_tokens=True, clean_up_tokenization_spaces=False)[0]
+        "What is your favorite condiment?"
+        ```"""
+        output_attentions = output_attentions if output_attentions is not None else self.config.output_attentions
+        output_hidden_states = (
+            output_hidden_states if output_hidden_states is not None else self.config.output_hidden_states
+        )
+
+        # decoder outputs consists of (dec_features, layer_state, dec_hidden, dec_attn)
+        outputs: BaseModelOutputWithPast = self.model(
+            input_ids=input_ids,
+            attention_mask=attention_mask,
+            position_ids=position_ids,
+            past_key_values=past_key_values,
+            inputs_embeds=inputs_embeds,
+            use_cache=use_cache,
+            output_attentions=output_attentions,
+            output_hidden_states=output_hidden_states,
+            cache_position=cache_position,
+            adarms_cond=adarms_cond,
+            **kwargs,
+        )
+
+        hidden_states = outputs.last_hidden_state
+        # Only compute necessary logits, and do not upcast them to float if we are not computing the loss
+        slice_indices = slice(-logits_to_keep, None) if isinstance(logits_to_keep, int) else logits_to_keep
+        logits = self.lm_head(hidden_states[:, slice_indices, :])
+
+        loss = None
+        if labels is not None:
+            loss = self.loss_function(logits=logits, labels=labels, vocab_size=self.config.vocab_size, **kwargs)
+
+        return CausalLMOutputWithPast(
+            loss=loss,
+            logits=logits,
+            past_key_values=outputs.past_key_values,
+            hidden_states=outputs.hidden_states,
+            attentions=outputs.attentions,
+        )
+
+
+@auto_docstring(
+    custom_intro="""
+    The Gemma Model transformer with a sequence classification head on top (linear layer).
+
+    [`GemmaForSequenceClassification`] uses the last token in order to do the classification, as other causal models
+    (e.g. GPT-2) do.
+
+    Since it does classification on the last token, it requires to know the position of the last token. If a
+    `pad_token_id` is defined in the configuration, it finds the last token that is not a padding token in each row. If
+    no `pad_token_id` is defined, it simply takes the last value in each row of the batch. Since it cannot guess the
+    padding tokens when `inputs_embeds` are passed instead of `input_ids`, it does the same (take the last value in
+    each row of the batch).
+    """
+)
+class GemmaForSequenceClassification(GemmaPreTrainedModel):
+    def __init__(self, config):
+        super().__init__(config)
+        self.num_labels = config.num_labels
+        self.model = GemmaModel(config)
+        self.score = nn.Linear(config.hidden_size, self.num_labels, bias=False)
+
+        # Initialize weights and apply final processing
+        self.post_init()
+
+    def get_input_embeddings(self):
+        return self.model.embed_tokens
+
+    def set_input_embeddings(self, value):
+        self.model.embed_tokens = value
+
+    @can_return_tuple
+    @auto_docstring
+    def forward(
+        self,
+        input_ids: Optional[torch.LongTensor] = None,
+        attention_mask: Optional[torch.Tensor] = None,
+        position_ids: Optional[torch.LongTensor] = None,
+        past_key_values: Optional[Cache] = None,
+        inputs_embeds: Optional[torch.FloatTensor] = None,
+        labels: Optional[torch.LongTensor] = None,
+        use_cache: Optional[bool] = None,
+        output_attentions: Optional[bool] = None,
+        output_hidden_states: Optional[bool] = None,
+        adarms_cond: Optional[torch.Tensor] = None,
+    ) -> SequenceClassifierOutputWithPast:
+        r"""
+        labels (`torch.LongTensor` of shape `(batch_size,)`, *optional*):
+            Labels for computing the sequence classification/regression loss. Indices should be in `[0, ...,
+            config.num_labels - 1]`. If `config.num_labels == 1` a regression loss is computed (Mean-Square loss), If
+            `config.num_labels > 1` a classification loss is computed (Cross-Entropy).
+
+        adarms_cond (`torch.Tensor` of shape `(batch_size, cond_dim)`, *optional*):
+            Condition for ADARMS.
+        """
+
+        transformer_outputs: BaseModelOutputWithPast = self.model(
+            input_ids,
+            attention_mask=attention_mask,
+            position_ids=position_ids,
+            past_key_values=past_key_values,
+            inputs_embeds=inputs_embeds,
+            use_cache=use_cache,
+            output_attentions=output_attentions,
+            output_hidden_states=output_hidden_states,
+            adarms_cond=adarms_cond,
+        )
+        hidden_states = transformer_outputs.last_hidden_state
+        logits = self.score(hidden_states)
+
+        if input_ids is not None:
+            batch_size = input_ids.shape[0]
+        else:
+            batch_size = inputs_embeds.shape[0]
+
+        if self.config.pad_token_id is None and batch_size != 1:
+            raise ValueError("Cannot handle batch sizes > 1 if no padding token is defined.")
+        if self.config.pad_token_id is None:
+            last_non_pad_token = -1
+        elif input_ids is not None:
+            # To handle both left- and right- padding, we take the rightmost token that is not equal to pad_token_id
+            non_pad_mask = (input_ids != self.config.pad_token_id).to(logits.device, torch.int32)
+            token_indices = torch.arange(input_ids.shape[-1], device=logits.device, dtype=torch.int32)
+            last_non_pad_token = (token_indices * non_pad_mask).argmax(-1)
+        else:
+            last_non_pad_token = -1
+            logger.warning_once(
+                f"{self.__class__.__name__} will not detect padding tokens in `inputs_embeds`. Results may be "
+                "unexpected if using padding tokens in conjunction with `inputs_embeds.`"
+            )
+
+        pooled_logits = logits[torch.arange(batch_size, device=logits.device), last_non_pad_token]
+
+        loss = None
+        if labels is not None:
+            loss = self.loss_function(logits=logits, labels=labels, pooled_logits=pooled_logits, config=self.config)
+
+        return SequenceClassifierOutputWithPast(
+            loss=loss,
+            logits=pooled_logits,
+            past_key_values=transformer_outputs.past_key_values,
+            hidden_states=transformer_outputs.hidden_states,
+            attentions=transformer_outputs.attentions,
+        )
+
+
+@auto_docstring
+class GemmaForTokenClassification(GemmaPreTrainedModel):
+    def __init__(self, config):
+        super().__init__(config)
+        self.num_labels = config.num_labels
+        self.model = GemmaModel(config)
+        if getattr(config, "classifier_dropout", None) is not None:
+            classifier_dropout = config.classifier_dropout
+        elif getattr(config, "hidden_dropout", None) is not None:
+            classifier_dropout = config.hidden_dropout
+        else:
+            classifier_dropout = 0.1
+        self.dropout = nn.Dropout(classifier_dropout)
+        self.score = nn.Linear(config.hidden_size, config.num_labels)
+
+        # Initialize weights and apply final processing
+        self.post_init()
+
+    def get_input_embeddings(self):
+        return self.model.embed_tokens
+
+    def set_input_embeddings(self, value):
+        self.model.embed_tokens = value
+
+    @can_return_tuple
+    @auto_docstring
+    def forward(
+        self,
+        input_ids: Optional[torch.LongTensor] = None,
+        attention_mask: Optional[torch.Tensor] = None,
+        position_ids: Optional[torch.LongTensor] = None,
+        past_key_values: Optional[Cache] = None,
+        inputs_embeds: Optional[torch.FloatTensor] = None,
+        labels: Optional[torch.LongTensor] = None,
+        use_cache: Optional[bool] = None,
+        output_attentions: Optional[bool] = None,
+        output_hidden_states: Optional[bool] = None,
+        adarms_cond: Optional[torch.Tensor] = None,
+    ) -> TokenClassifierOutput:
+        r"""
+        labels (`torch.LongTensor` of shape `(batch_size,)`, *optional*):
+            Labels for computing the sequence classification/regression loss. Indices should be in `[0, ...,
+            config.num_labels - 1]`. If `config.num_labels == 1` a regression loss is computed (Mean-Square loss), If
+            `config.num_labels > 1` a classification loss is computed (Cross-Entropy).
+
+        adarms_cond (`torch.Tensor` of shape `(batch_size, cond_dim)`, *optional*):
+            Condition for ADARMS.
+        """
+
+        outputs: BaseModelOutputWithPast = self.model(
+            input_ids,
+            attention_mask=attention_mask,
+            position_ids=position_ids,
+            past_key_values=past_key_values,
+            inputs_embeds=inputs_embeds,
+            use_cache=use_cache,
+            output_attentions=output_attentions,
+            output_hidden_states=output_hidden_states,
+            adarms_cond=adarms_cond,
+        )
+        sequence_output = outputs.last_hidden_state
+        sequence_output = self.dropout(sequence_output)
+        logits = self.score(sequence_output)
+
+        loss = None
+        if labels is not None:
+            loss = self.loss_function(logits, labels, self.config)
+
+        return TokenClassifierOutput(
+            loss=loss,
+            logits=logits,
+            hidden_states=outputs.hidden_states,
+            attentions=outputs.attentions,
+        )
+
+
+__all__ = [
+    "GemmaModel",
+    "GemmaForCausalLM",
+    "GemmaForSequenceClassification",
+    "GemmaForTokenClassification",
+    "GemmaPreTrainedModel",
+]

real_inference/source/XPolicyLab/policy/Pi_05/openpi/src/openpi/models_pytorch/transformers_replace/models/paligemma/modeling_paligemma.py

@@ -0,0 +1,622 @@
+# coding=utf-8
+# Copyright 2024 the HuggingFace Inc. team. All rights reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+"""PyTorch PaliGemmamodel."""
+
+from dataclasses import dataclass
+from typing import Optional, Union
+
+import torch
+import torch.utils.checkpoint
+from torch import nn
+
+from ...cache_utils import Cache, HybridCache, StaticCache
+from ...generation import GenerationMixin
+from ...modeling_flash_attention_utils import FlashAttentionKwargs
+from ...modeling_outputs import BaseModelOutputWithPast
+from ...modeling_utils import PreTrainedModel
+from ...processing_utils import Unpack
+from ...utils import LossKwargs, ModelOutput, auto_docstring, can_return_tuple, is_torchdynamo_compiling, logging
+from ..auto import AutoModel
+from .configuration_paligemma import PaliGemmaConfig
+
+
+logger = logging.get_logger(__name__)
+
+
+@dataclass
+@auto_docstring(
+    custom_intro="""
+    Base class for Paligemma outputs, with hidden states and attentions.
+    """
+)
+class PaligemmaModelOutputWithPast(BaseModelOutputWithPast):
+    r"""
+    past_key_values (`tuple(tuple(torch.FloatTensor))`, *optional*, returned when `use_cache=True` is passed or when `config.use_cache=True`):
+        Tuple of `tuple(torch.FloatTensor)` of length `config.n_layers`, with each tuple having 2 tensors of shape
+        `(batch_size, num_heads, sequence_length, embed_size_per_head)`)
+
+        Contains pre-computed hidden-states (key and values in the self-attention blocks) that can be used (see
+        `past_key_values` input) to speed up sequential decoding.
+    image_hidden_states (`torch.FloatTensor`, *optional*):
+        A `torch.FloatTensor` of size `(batch_size, num_images, sequence_length, hidden_size)`.
+        image_hidden_states of the model produced by the vision encoder and after projecting the last hidden state.
+    """
+
+    image_hidden_states: Optional[torch.FloatTensor] = None
+
+
+@dataclass
+@auto_docstring(
+    custom_intro="""
+    Base class for PaliGemma causal language model (or autoregressive) outputs.
+    """
+)
+class PaliGemmaCausalLMOutputWithPast(ModelOutput):
+    r"""
+    loss (`torch.FloatTensor` of shape `(1,)`, *optional*, returned when `labels` is provided):
+        Language modeling loss (for next-token prediction).
+    logits (`torch.FloatTensor` of shape `(batch_size, sequence_length, config.text_config.vocab_size)`):
+        Prediction scores of the language modeling head (scores for each vocabulary token before SoftMax).
+    past_key_values (`tuple(tuple(torch.FloatTensor))`, *optional*, returned when `use_cache=True` is passed or when `config.use_cache=True`):
+        Tuple of `tuple(torch.FloatTensor)` of length `config.n_layers`, with each tuple having 2 tensors of shape
+        `(batch_size, num_heads, sequence_length, embed_size_per_head)`)
+
+        Contains pre-computed hidden-states (key and values in the self-attention blocks) that can be used (see
+        `past_key_values` input) to speed up sequential decoding.
+    image_hidden_states (`torch.FloatTensor`, *optional*):
+        A `torch.FloatTensor` of size `(batch_size, num_images, sequence_length, hidden_size)`.
+        image_hidden_states of the model produced by the vision encoder after projecting last hidden state.
+    """
+
+    loss: Optional[torch.FloatTensor] = None
+    logits: Optional[torch.FloatTensor] = None
+    past_key_values: Optional[Union[list[torch.FloatTensor], Cache]] = None
+    hidden_states: Optional[tuple[torch.FloatTensor]] = None
+    attentions: Optional[tuple[torch.FloatTensor]] = None
+    image_hidden_states: Optional[torch.FloatTensor] = None
+
+
+class PaliGemmaMultiModalProjector(nn.Module):
+    def __init__(self, config: PaliGemmaConfig):
+        super().__init__()
+        self.linear = nn.Linear(config.vision_config.hidden_size, config.vision_config.projection_dim, bias=True)
+
+    def forward(self, image_features):
+        hidden_states = self.linear(image_features)
+
+        return hidden_states
+
+
+@auto_docstring
+class PaliGemmaPreTrainedModel(PreTrainedModel):
+    config_class = PaliGemmaConfig
+    base_model_prefix = ""
+    supports_gradient_checkpointing = True
+    _no_split_modules = ["PaliGemmaMultiModalProjector"]
+    _skip_keys_device_placement = "past_key_values"
+    _supports_cache_class = True
+    _supports_quantized_cache = True
+    _supports_static_cache = True
+    _supports_flash_attn_2 = True
+    _supports_sdpa = True
+    _supports_flex_attn = True
+    _supports_attention_backend = True
+
+    def _init_weights(self, module):
+        # important: this ported version of PaliGemmaisn't meant for training from scratch - only
+        # inference and fine-tuning
+        std = getattr(self.config, "initializer_range", self.config.get_text_config().initializer_range)
+
+        if isinstance(module, nn.Linear):
+            module.weight.data.normal_(mean=0.0, std=std)
+            if module.bias is not None:
+                module.bias.data.zero_()
+
+
+@auto_docstring(
+    custom_intro="""
+    The Base Paligemma model which consists of a vision backbone and a language model withou language modeling head.,
+    """
+)
+class PaliGemmaModel(PaliGemmaPreTrainedModel):
+    _checkpoint_conversion_mapping = {"language_model.model": "language_model"}
+    # we are filtering the logits/labels so we shouldn't divide the loss based on num_items_in_batch
+    accepts_loss_kwargs = False
+
+    def __init__(self, config: PaliGemmaConfig):
+        super().__init__(config)
+        self.vision_tower = AutoModel.from_config(config=config.vision_config)
+        self.multi_modal_projector = PaliGemmaMultiModalProjector(config)
+        self.vocab_size = config.text_config.vocab_size
+
+        language_model = AutoModel.from_config(config=config.text_config)
+        self.language_model = language_model
+
+        self.pad_token_id = self.config.pad_token_id if self.config.pad_token_id is not None else -1
+        self.post_init()
+
+    # Copied from transformers.models.llava.modeling_llava.LlavaModel.get_input_embeddings with Llava->PaliGemma
+    def get_input_embeddings(self):
+        return self.language_model.get_input_embeddings()
+
+    # Copied from transformers.models.llava.modeling_llava.LlavaModel.set_input_embeddings with Llava->PaliGemma
+    def set_input_embeddings(self, value):
+        self.language_model.set_input_embeddings(value)
+
+    def set_decoder(self, decoder):
+        self.language_model = decoder
+
+    def get_decoder(self):
+        return self.language_model
+
+    def _update_causal_mask(
+        self,
+        attention_mask,
+        token_type_ids=None,
+        past_key_values=None,
+        cache_position=None,
+        input_tensor=None,
+        is_training: Optional[bool] = None,
+    ):
+        if self.config.text_config._attn_implementation == "flash_attention_2":
+            if attention_mask is not None and 0.0 in attention_mask:
+                return attention_mask
+            return None
+        is_training = is_training if is_training is not None else self.training
+        using_static_cache = isinstance(past_key_values, StaticCache)
+        min_dtype = torch.finfo(self.dtype).min
+        if input_tensor is None:
+            input_tensor = attention_mask
+
+        inputs_lead_dim, sequence_length = input_tensor.shape[:2]
+        if using_static_cache:
+            target_length = past_key_values.get_max_cache_shape()
+        elif isinstance(past_key_values, HybridCache):
+            target_length = past_key_values.get_max_cache_shape()
+        else:
+            target_length = (
+                attention_mask.shape[-1]
+                if isinstance(attention_mask, torch.Tensor)
+                else cache_position[0] + sequence_length + 1
+            )
+
+        if attention_mask is not None and attention_mask.dim() == 4:
+            # In this case we assume that the mask comes already in inverted form and requires no inversion or slicing.
+            return attention_mask
+
+        causal_mask = torch.full(
+            (sequence_length, target_length), fill_value=min_dtype, dtype=self.dtype, device=cache_position.device
+        )
+        # Causal diagonal mask only if training, otherwise attend to the whole prefix. Training-specific attn for prefix is handled below
+        if sequence_length != 1:
+            if is_training:
+                causal_mask = torch.triu(causal_mask, diagonal=1)
+            else:
+                causal_mask[:, :sequence_length] = 0.0
+
+        causal_mask *= torch.arange(target_length, device=cache_position.device) > cache_position.reshape(-1, 1)
+        causal_mask = causal_mask[None, None, :, :].expand(inputs_lead_dim, 1, -1, -1)
+        if attention_mask is not None:
+            causal_mask = causal_mask.clone()  # copy to contiguous memory for in-place edit
+            mask_length = attention_mask.shape[-1]
+
+            # First unmask prefix tokens during training
+            if is_training:
+                if token_type_ids is None:
+                    raise ValueError("Token type ids must be provided during training")
+                causal_mask[:, :, :, :mask_length] = causal_mask[:, :, :, :mask_length].masked_fill(
+                    token_type_ids[:, None, None, :].to(causal_mask.device) == 0, 0
+                )
+
+            # Then apply padding mask (will mask pad tokens)
+            padding_mask = causal_mask[:, :, :, :mask_length] + attention_mask[:, None, None, :].to(causal_mask.device)
+            padding_mask = padding_mask == 0
+            causal_mask[:, :, :, :mask_length] = causal_mask[:, :, :, :mask_length].masked_fill(
+                padding_mask, min_dtype
+            )
+
+        return causal_mask
+
+    def get_image_features(self, pixel_values: torch.FloatTensor):
+        """
+        Obtains image last hidden states from the vision tower and apply multimodal projection.
+
+        Args:
+            pixel_values (`torch.FloatTensor]` of shape `(batch_size, channels, height, width)`)
+               The tensors corresponding to the input images.
+        Returns:
+            image_features (`torch.Tensor`): Image feature tensor of shape `(num_images, image_length, embed_dim)`).
+        """
+        image_outputs = self.vision_tower(pixel_values)
+        selected_image_feature = image_outputs.last_hidden_state
+        image_features = self.multi_modal_projector(selected_image_feature)
+        return image_features
+
+    @can_return_tuple
+    @auto_docstring
+    def forward(
+        self,
+        input_ids: torch.LongTensor = None,
+        pixel_values: torch.FloatTensor = None,
+        attention_mask: Optional[torch.Tensor] = None,
+        position_ids: Optional[torch.LongTensor] = None,
+        past_key_values: Optional[Union[list[torch.FloatTensor], Cache]] = None,
+        token_type_ids: Optional[torch.LongTensor] = None,
+        cache_position: Optional[torch.LongTensor] = None,
+        inputs_embeds: Optional[torch.FloatTensor] = None,
+        labels: Optional[torch.LongTensor] = None,
+        use_cache: Optional[bool] = None,
+        output_attentions: Optional[bool] = None,
+        output_hidden_states: Optional[bool] = None,
+        return_dict: Optional[bool] = None,
+        **kwargs: Unpack[FlashAttentionKwargs],
+    ) -> Union[tuple, PaligemmaModelOutputWithPast]:
+        r"""
+        labels (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*):
+            Labels for computing the masked language modeling loss. Indices should either be in `[0, ...,
+            config.text_config.vocab_size]` or -100 (see `input_ids` docstring). Tokens with indices set to `-100` are ignored
+            (masked), the loss is only computed for the tokens with labels in `[0, ..., config.text_config.vocab_size]`.
+
+        Example:
+
+        ```python
+        >>> from PIL import Image
+        >>> import requests
+        >>> from transformers import AutoProcessor, PaliGemmaForConditionalGeneration
+
+        >>> model = PaliGemmaForConditionalGeneration.from_pretrained("google/paligemma2-3b-mix-224")
+        >>> processor = AutoProcessor.from_pretrained("google/paligemma2-3b-mix-224")
+
+        >>> prompt = "Where is the cat standing?"
+        >>> url = "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/pipeline-cat-chonk.jpeg"
+        >>> image = Image.open(requests.get(url, stream=True).raw)
+
+        >>> inputs = processor(images=image, text=prompt,  return_tensors="pt")
+
+        >>> # Generate
+        >>> generate_ids = model.generate(**inputs,)
+        >>> processor.batch_decode(generate_ids, skip_special_tokens=True, clean_up_tokenization_spaces=False)[0]
+        "Where is the cat standing?\nsnow"
+        ```"""
+
+        if (input_ids is None) ^ (inputs_embeds is not None):
+            raise ValueError("You must specify exactly one of input_ids or inputs_embeds")
+
+        output_attentions = output_attentions if output_attentions is not None else self.config.output_attentions
+        output_hidden_states = (
+            output_hidden_states if output_hidden_states is not None else self.config.output_hidden_states
+        )
+        return_dict = return_dict if return_dict is not None else self.config.use_return_dict
+
+        is_training = token_type_ids is not None and labels is not None
+
+        # Replace image id woth PAD if the image token if OOV, to avoid index-errors
+        if input_ids is not None and self.config.image_token_id >= self.vocab_size:
+            special_image_mask = input_ids == self.config.image_token_id
+            llm_input_ids = input_ids.clone()
+            llm_input_ids[special_image_mask] = 0
+        else:
+            llm_input_ids = input_ids
+
+        if inputs_embeds is None:
+            inputs_embeds = self.get_input_embeddings()(llm_input_ids)
+
+        if cache_position is None:
+            past_seen_tokens = past_key_values.get_seq_length() if past_key_values is not None else 0
+            cache_position = torch.arange(
+                past_seen_tokens, past_seen_tokens + inputs_embeds.shape[1], device=inputs_embeds.device
+            )
+
+        if position_ids is None:
+            position_ids = cache_position.unsqueeze(0) + 1  # Paligemma positions are 1-indexed
+
+        # Merge text and images
+        if pixel_values is not None:
+            image_features = self.get_image_features(pixel_values)
+
+            if input_ids is None:
+                special_image_mask = inputs_embeds == self.get_input_embeddings()(
+                    torch.tensor(self.config.image_token_id, dtype=torch.long, device=inputs_embeds.device)
+                )
+            else:
+                special_image_mask = (input_ids == self.config.image_token_id).unsqueeze(-1)
+                special_image_mask = special_image_mask.expand_as(inputs_embeds).to(inputs_embeds.device)
+
+            if not is_torchdynamo_compiling() and inputs_embeds[special_image_mask].numel() != image_features.numel():
+                image_tokens_in_text = (special_image_mask).sum(dim=1).sum(dim=0)[0]
+                raise ValueError(
+                    f"Number of images does not match number of special image tokens in the input text. "
+                    f"Got {image_tokens_in_text} image tokens in the text but {image_features.shape[0] * image_features.shape[1]} "
+                    "tokens from image embeddings."
+                )
+            image_features = image_features.to(inputs_embeds.device, inputs_embeds.dtype)
+            inputs_embeds = inputs_embeds.masked_scatter(special_image_mask, image_features)
+
+        causal_mask = self._update_causal_mask(
+            attention_mask, token_type_ids, past_key_values, cache_position, inputs_embeds, is_training
+        )
+        outputs = self.language_model(
+            attention_mask=causal_mask,
+            position_ids=position_ids,
+            past_key_values=past_key_values,
+            inputs_embeds=inputs_embeds,
+            use_cache=use_cache,
+            output_attentions=output_attentions,
+            output_hidden_states=output_hidden_states,
+            return_dict=True,
+            cache_position=cache_position,
+            **kwargs,
+        )
+
+        return PaligemmaModelOutputWithPast(
+            last_hidden_state=outputs.last_hidden_state,
+            past_key_values=outputs.past_key_values,
+            hidden_states=outputs.hidden_states,
+            attentions=outputs.attentions,
+            image_hidden_states=image_features if pixel_values is not None else None,
+        )
+
+
+class KwargsForCausalLM(FlashAttentionKwargs, LossKwargs): ...
+
+
+@auto_docstring(
+    custom_intro="""
+    The Base Paligemma model which consists of a vision backbone and a language model without language modeling head.,
+    """
+)
+class PaliGemmaForConditionalGeneration(PaliGemmaPreTrainedModel, GenerationMixin):
+    _checkpoint_conversion_mapping = {
+        "^language_model.model": "model.language_model",
+        "^vision_tower": "model.vision_tower",
+        "^multi_modal_projector": "model.multi_modal_projector",
+        "^language_model.lm_head": "lm_head",
+    }
+    _tied_weights_keys = ["lm_head.weight"]
+
+    def __init__(self, config: PaliGemmaConfig):
+        super().__init__(config)
+        self.model = PaliGemmaModel(config)
+        self.lm_head = nn.Linear(config.text_config.hidden_size, config.text_config.vocab_size, bias=False)
+        self.post_init()
+
+    def get_input_embeddings(self):
+        return self.model.get_input_embeddings()
+
+    def set_input_embeddings(self, value):
+        self.model.set_input_embeddings(value)
+
+    def get_output_embeddings(self):
+        return self.lm_head
+
+    def set_output_embeddings(self, new_embeddings):
+        self.lm_head = new_embeddings
+
+    def set_decoder(self, decoder):
+        self.model.set_decoder(decoder)
+
+    def get_decoder(self):
+        return self.model.get_decoder()
+
+    def get_image_features(self, pixel_values):
+        return self.model.get_image_features(pixel_values)
+
+    # Make modules available throught conditional class for BC
+    @property
+    def language_model(self):
+        return self.model.language_model
+
+    @property
+    def vision_tower(self):
+        return self.model.vision_tower
+
+    @property
+    def multi_modal_projector(self):
+        return self.model.multi_modal_projector
+
+    @can_return_tuple
+    @auto_docstring
+    def forward(
+        self,
+        input_ids: torch.LongTensor = None,
+        pixel_values: torch.FloatTensor = None,
+        attention_mask: Optional[torch.Tensor] = None,
+        position_ids: Optional[torch.LongTensor] = None,
+        past_key_values: Optional[Union[list[torch.FloatTensor], Cache]] = None,
+        token_type_ids: Optional[torch.LongTensor] = None,
+        cache_position: Optional[torch.LongTensor] = None,
+        inputs_embeds: Optional[torch.FloatTensor] = None,
+        labels: Optional[torch.LongTensor] = None,
+        use_cache: Optional[bool] = None,
+        output_attentions: Optional[bool] = None,
+        output_hidden_states: Optional[bool] = None,
+        return_dict: Optional[bool] = None,
+        logits_to_keep: Union[int, torch.Tensor] = 0,
+        **kwargs: Unpack[KwargsForCausalLM],
+    ) -> Union[tuple, PaliGemmaCausalLMOutputWithPast]:
+        r"""
+        labels (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*):
+            Labels for computing the masked language modeling loss. Indices should either be in `[0, ...,
+            config.text_config.vocab_size]` or -100 (see `input_ids` docstring). Tokens with indices set to `-100` are ignored
+            (masked), the loss is only computed for the tokens with labels in `[0, ..., config.text_config.vocab_size]`.
+
+        Example:
+
+        ```python
+        >>> from PIL import Image
+        >>> import requests
+        >>> from transformers import AutoProcessor, PaliGemmaForConditionalGeneration
+
+        >>> model = PaliGemmaForConditionalGeneration.from_pretrained("google/paligemma2-3b-mix-224")
+        >>> processor = AutoProcessor.from_pretrained("google/paligemma2-3b-mix-224")
+
+        >>> prompt = "Where is the cat standing?"
+        >>> url = "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/pipeline-cat-chonk.jpeg"
+        >>> image = Image.open(requests.get(url, stream=True).raw)
+
+        >>> inputs = processor(images=image, text=prompt,  return_tensors="pt")
+
+        >>> # Generate
+        >>> generate_ids = model.generate(**inputs,)
+        >>> processor.batch_decode(generate_ids, skip_special_tokens=True, clean_up_tokenization_spaces=False)[0]
+        "Where is the cat standing?\nsnow"
+        ```"""
+        output_attentions = output_attentions if output_attentions is not None else self.config.output_attentions
+        output_hidden_states = (
+            output_hidden_states if output_hidden_states is not None else self.config.output_hidden_states
+        )
+        return_dict = return_dict if return_dict is not None else self.config.use_return_dict
+
+        outputs = self.model(
+            input_ids=input_ids,
+            pixel_values=pixel_values,
+            token_type_ids=token_type_ids,
+            attention_mask=attention_mask,
+            position_ids=position_ids,
+            past_key_values=past_key_values,
+            inputs_embeds=inputs_embeds,
+            use_cache=use_cache,
+            labels=labels,
+            output_attentions=output_attentions,
+            output_hidden_states=output_hidden_states,
+            return_dict=True,
+            cache_position=cache_position,
+            **kwargs,
+        )
+
+        hidden_states = outputs[0]
+        # Only compute necessary logits, and do not upcast them to float if we are not computing the loss
+        slice_indices = slice(-logits_to_keep, None) if isinstance(logits_to_keep, int) else logits_to_keep
+        logits = self.lm_head(hidden_states[:, slice_indices, :])
+
+        loss = None
+        if labels is not None:
+            loss = self.loss_function(
+                logits=logits, labels=labels, vocab_size=self.config.text_config.vocab_size, **kwargs
+            )
+
+        return PaliGemmaCausalLMOutputWithPast(
+            loss=loss,
+            logits=logits,
+            past_key_values=outputs.past_key_values,
+            hidden_states=outputs.hidden_states,
+            attentions=outputs.attentions,
+            image_hidden_states=outputs.image_hidden_states,
+        )
+
+    def prepare_inputs_for_generation(
+        self,
+        input_ids,
+        past_key_values=None,
+        inputs_embeds=None,
+        cache_position=None,
+        position_ids=None,
+        pixel_values=None,
+        attention_mask=None,
+        token_type_ids=None,
+        use_cache=True,
+        logits_to_keep=None,
+        labels=None,
+        **kwargs,
+    ):
+        # Overwritten -- custom `position_ids` and `pixel_values` handling
+        model_inputs = super().prepare_inputs_for_generation(
+            input_ids,
+            past_key_values=past_key_values,
+            inputs_embeds=inputs_embeds,
+            attention_mask=attention_mask,
+            position_ids=position_ids,
+            cache_position=cache_position,
+            use_cache=use_cache,
+            logits_to_keep=logits_to_keep,
+            token_type_ids=token_type_ids,
+            **kwargs,
+        )
+
+        # position_ids in Paligemma are 1-indexed
+        if model_inputs.get("position_ids") is not None:
+            model_inputs["position_ids"] += 1
+        # If we're in cached decoding stage, pixel values should be None because input ids do not contain special image token anymore
+        # Otherwise we need pixel values to be passed to model. NOTE: use_cache=False needs pixel_values always
+        if cache_position[0] == 0:
+            model_inputs["pixel_values"] = pixel_values
+        is_training = token_type_ids is not None and labels is not None
+        if cache_position[0] == 0 and isinstance(past_key_values, HybridCache):
+            input_tensor = inputs_embeds if inputs_embeds is not None else input_ids
+            causal_mask = self.model._update_causal_mask(
+                attention_mask, token_type_ids, past_key_values, cache_position, input_tensor, is_training
+            )
+            model_inputs["attention_mask"] = causal_mask
+
+        return model_inputs
+
+    @staticmethod
+    # Copied from transformers.models.gptj.modeling_gptj.GPTJModel._prepare_4d_causal_attention_mask_with_cache_position
+    def _prepare_4d_causal_attention_mask_with_cache_position(
+        attention_mask: torch.Tensor,
+        sequence_length: int,
+        target_length: int,
+        dtype: torch.dtype,
+        cache_position: torch.Tensor,
+        batch_size: int,
+        **kwargs,
+    ):
+        """
+        Creates a causal 4D mask of shape `(batch_size, 1, query_length, key_value_length)` from a 2D mask of shape
+        `(batch_size, key_value_length)`, or if the input `attention_mask` is already 4D, do nothing.
+
+        Args:
+            attention_mask (`torch.Tensor`):
+                A 2D attention mask of shape `(batch_size, key_value_length)` or a 4D attention mask of shape
+                `(batch_size, 1, query_length, key_value_length)`.
+            sequence_length (`int`):
+                The sequence length being processed.
+            target_length (`int`):
+                The target length: when generating with static cache, the mask should be as long as the static cache,
+                to account for the 0 padding, the part of the cache that is not filled yet.
+            dtype (`torch.dtype`):
+                The dtype to use for the 4D attention mask.
+            cache_position (`torch.Tensor`):
+                Indices depicting the position of the input sequence tokens in the sequence.
+            batch_size (`torch.Tensor`):
+                Batch size.
+        """
+        if attention_mask is not None and attention_mask.dim() == 4:
+            # In this case we assume that the mask comes already in inverted form and requires no inversion or slicing.
+            causal_mask = attention_mask
+        else:
+            min_dtype = torch.finfo(dtype).min
+            causal_mask = torch.full(
+                (sequence_length, target_length), fill_value=min_dtype, dtype=dtype, device=cache_position.device
+            )
+            if sequence_length != 1:
+                causal_mask = torch.triu(causal_mask, diagonal=1)
+            causal_mask *= torch.arange(target_length, device=cache_position.device) > cache_position.reshape(-1, 1)
+            causal_mask = causal_mask[None, None, :, :].expand(batch_size, 1, -1, -1)
+            if attention_mask is not None:
+                causal_mask = causal_mask.clone()  # copy to contiguous memory for in-place edit
+                mask_length = attention_mask.shape[-1]
+                padding_mask = causal_mask[:, :, :, :mask_length] + attention_mask[:, None, None, :].to(
+                    causal_mask.device
+                )
+                padding_mask = padding_mask == 0
+                causal_mask[:, :, :, :mask_length] = causal_mask[:, :, :, :mask_length].masked_fill(
+                    padding_mask, min_dtype
+                )
+
+        return causal_mask
+
+
+__all__ = ["PaliGemmaForConditionalGeneration", "PaliGemmaPreTrainedModel", "PaliGemmaModel"]

real_inference/source/XPolicyLab/policy/Pi_05/openpi/src/openpi/models_pytorch/transformers_replace/models/siglip/check.py

@@ -0,0 +1,4 @@
+import transformers
+
+def check_whether_transformers_replace_is_installed_correctly():
+    return transformers.__version__ == "4.53.2"

real_inference/source/XPolicyLab/policy/Pi_05/openpi/src/openpi/models_pytorch/transformers_replace/models/siglip/modeling_siglip.py

@@ -0,0 +1,1237 @@
+# coding=utf-8
+# Copyright 2024 Google AI and The HuggingFace Team. All rights reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+"""PyTorch Siglip model."""
+
+import math
+import warnings
+from dataclasses import dataclass
+from typing import Any, Callable, Optional, Union
+
+import numpy as np
+import torch
+import torch.utils.checkpoint
+from torch import nn
+from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss, MSELoss
+from torch.nn.init import _calculate_fan_in_and_fan_out
+
+from ...activations import ACT2FN
+from ...modeling_attn_mask_utils import _prepare_4d_attention_mask
+from ...modeling_layers import GradientCheckpointingLayer
+from ...modeling_outputs import BaseModelOutput, BaseModelOutputWithPooling, ImageClassifierOutput
+from ...modeling_utils import ALL_ATTENTION_FUNCTIONS, PreTrainedModel
+from ...utils import ModelOutput, auto_docstring, can_return_tuple, logging, torch_int
+from .configuration_siglip import SiglipConfig, SiglipTextConfig, SiglipVisionConfig
+
+
+logger = logging.get_logger(__name__)
+
+
+def _trunc_normal_(tensor, mean, std, a, b):
+    # Cut & paste from PyTorch official master until it's in a few official releases - RW
+    # Method based on https://people.sc.fsu.edu/~jburkardt/presentations/truncated_normal.pdf
+    def norm_cdf(x):
+        # Computes standard normal cumulative distribution function
+        return (1.0 + math.erf(x / math.sqrt(2.0))) / 2.0
+
+    if (mean < a - 2 * std) or (mean > b + 2 * std):
+        warnings.warn(
+            "mean is more than 2 std from [a, b] in nn.init.trunc_normal_. "
+            "The distribution of values may be incorrect.",
+            stacklevel=2,
+        )
+
+    # Values are generated by using a truncated uniform distribution and
+    # then using the inverse CDF for the normal distribution.
+    # Get upper and lower cdf values
+    l = norm_cdf((a - mean) / std)
+    u = norm_cdf((b - mean) / std)
+
+    # Uniformly fill tensor with values from [l, u], then translate to
+    # [2l-1, 2u-1].
+    tensor.uniform_(2 * l - 1, 2 * u - 1)
+
+    # Use inverse cdf transform for normal distribution to get truncated
+    # standard normal
+    tensor.erfinv_()
+
+    # Transform to proper mean, std
+    tensor.mul_(std * math.sqrt(2.0))
+    tensor.add_(mean)
+
+    # Clamp to ensure it's in the proper range
+    tensor.clamp_(min=a, max=b)
+
+
+def trunc_normal_tf_(
+    tensor: torch.Tensor, mean: float = 0.0, std: float = 1.0, a: float = -2.0, b: float = 2.0
+) -> torch.Tensor:
+    """Fills the input Tensor with values drawn from a truncated
+    normal distribution. The values are effectively drawn from the
+    normal distribution :math:`\\mathcal{N}(\text{mean}, \text{std}^2)`
+    with values outside :math:`[a, b]` redrawn until they are within
+    the bounds. The method used for generating the random values works
+    best when :math:`a \\leq \text{mean} \\leq b`.
+
+    NOTE: this 'tf' variant behaves closer to Tensorflow / JAX impl where the
+    bounds [a, b] are applied when sampling the normal distribution with mean=0, std=1.0
+    and the result is subsequently scaled and shifted by the mean and std args.
+
+    Args:
+        tensor: an n-dimensional `torch.Tensor`
+        mean: the mean of the normal distribution
+        std: the standard deviation of the normal distribution
+        a: the minimum cutoff value
+        b: the maximum cutoff value
+    """
+    with torch.no_grad():
+        _trunc_normal_(tensor, 0, 1.0, a, b)
+        tensor.mul_(std).add_(mean)
+
+
+def variance_scaling_(tensor, scale=1.0, mode="fan_in", distribution="normal"):
+    fan_in, fan_out = _calculate_fan_in_and_fan_out(tensor)
+    if mode == "fan_in":
+        denom = fan_in
+    elif mode == "fan_out":
+        denom = fan_out
+    elif mode == "fan_avg":
+        denom = (fan_in + fan_out) / 2
+
+    variance = scale / denom
+
+    if distribution == "truncated_normal":
+        # constant is stddev of standard normal truncated to (-2, 2)
+        trunc_normal_tf_(tensor, std=math.sqrt(variance) / 0.87962566103423978)
+    elif distribution == "normal":
+        with torch.no_grad():
+            tensor.normal_(std=math.sqrt(variance))
+    elif distribution == "uniform":
+        bound = math.sqrt(3 * variance)
+        with torch.no_grad():
+            tensor.uniform_(-bound, bound)
+    else:
+        raise ValueError(f"invalid distribution {distribution}")
+
+
+def lecun_normal_(tensor):
+    variance_scaling_(tensor, mode="fan_in", distribution="truncated_normal")
+
+
+def default_flax_embed_init(tensor):
+    variance_scaling_(tensor, mode="fan_in", distribution="normal")
+
+
+@dataclass
+@auto_docstring(
+    custom_intro="""
+    Base class for vision model's outputs that also contains image embeddings of the pooling of the last hidden states.
+    """
+)
+# Copied from transformers.models.clip.modeling_clip.CLIPVisionModelOutput with CLIP->Siglip
+class SiglipVisionModelOutput(ModelOutput):
+    r"""
+    image_embeds (`torch.FloatTensor` of shape `(batch_size, output_dim)` *optional* returned when model is initialized with `with_projection=True`):
+        The image embeddings obtained by applying the projection layer to the pooler_output.
+    """
+
+    image_embeds: Optional[torch.FloatTensor] = None
+    last_hidden_state: Optional[torch.FloatTensor] = None
+    hidden_states: Optional[tuple[torch.FloatTensor, ...]] = None
+    attentions: Optional[tuple[torch.FloatTensor, ...]] = None
+
+
+@dataclass
+@auto_docstring(
+    custom_intro="""
+    Base class for text model's outputs that also contains a pooling of the last hidden states.
+    """
+)
+# Copied from transformers.models.clip.modeling_clip.CLIPTextModelOutput with CLIP->Siglip
+class SiglipTextModelOutput(ModelOutput):
+    r"""
+    text_embeds (`torch.FloatTensor` of shape `(batch_size, output_dim)` *optional* returned when model is initialized with `with_projection=True`):
+        The text embeddings obtained by applying the projection layer to the pooler_output.
+    """
+
+    text_embeds: Optional[torch.FloatTensor] = None
+    last_hidden_state: Optional[torch.FloatTensor] = None
+    hidden_states: Optional[tuple[torch.FloatTensor, ...]] = None
+    attentions: Optional[tuple[torch.FloatTensor, ...]] = None
+
+
+@dataclass
+@auto_docstring
+# Copied from transformers.models.clip.modeling_clip.CLIPOutput with CLIP->Siglip
+class SiglipOutput(ModelOutput):
+    r"""
+    loss (`torch.FloatTensor` of shape `(1,)`, *optional*, returned when `return_loss` is `True`):
+        Contrastive loss for image-text similarity.
+    logits_per_image (`torch.FloatTensor` of shape `(image_batch_size, text_batch_size)`):
+        The scaled dot product scores between `image_embeds` and `text_embeds`. This represents the image-text
+        similarity scores.
+    logits_per_text (`torch.FloatTensor` of shape `(text_batch_size, image_batch_size)`):
+        The scaled dot product scores between `text_embeds` and `image_embeds`. This represents the text-image
+        similarity scores.
+    text_embeds (`torch.FloatTensor` of shape `(batch_size, output_dim`):
+        The text embeddings obtained by applying the projection layer to the pooled output of [`SiglipTextModel`].
+    image_embeds (`torch.FloatTensor` of shape `(batch_size, output_dim`):
+        The image embeddings obtained by applying the projection layer to the pooled output of [`SiglipVisionModel`].
+    text_model_output (`BaseModelOutputWithPooling`):
+        The output of the [`SiglipTextModel`].
+    vision_model_output (`BaseModelOutputWithPooling`):
+        The output of the [`SiglipVisionModel`].
+    """
+
+    loss: Optional[torch.FloatTensor] = None
+    logits_per_image: Optional[torch.FloatTensor] = None
+    logits_per_text: Optional[torch.FloatTensor] = None
+    text_embeds: Optional[torch.FloatTensor] = None
+    image_embeds: Optional[torch.FloatTensor] = None
+    text_model_output: BaseModelOutputWithPooling = None
+    vision_model_output: BaseModelOutputWithPooling = None
+
+    def to_tuple(self) -> tuple[Any]:
+        return tuple(
+            self[k] if k not in ["text_model_output", "vision_model_output"] else getattr(self, k).to_tuple()
+            for k in self.keys()
+        )
+
+
+class SiglipVisionEmbeddings(nn.Module):
+    def __init__(self, config: SiglipVisionConfig):
+        super().__init__()
+        self.config = config
+        self.embed_dim = config.hidden_size
+        self.image_size = config.image_size
+        self.patch_size = config.patch_size
+
+        self.patch_embedding = nn.Conv2d(
+            in_channels=config.num_channels,
+            out_channels=self.embed_dim,
+            kernel_size=self.patch_size,
+            stride=self.patch_size,
+            padding="valid",
+        )
+
+        self.num_patches = (self.image_size // self.patch_size) ** 2
+        self.num_positions = self.num_patches
+        self.position_embedding = nn.Embedding(self.num_positions, self.embed_dim)
+        self.register_buffer("position_ids", torch.arange(self.num_positions).expand((1, -1)), persistent=False)
+
+    def interpolate_pos_encoding(self, embeddings: torch.Tensor, height: int, width: int) -> torch.Tensor:
+        """
+        This method allows to interpolate the pre-trained position encodings, to be able to use the model on higher resolution
+        images. This method is also adapted to support torch.jit tracing and no class embeddings.
+
+        Adapted from:
+        - https://github.com/facebookresearch/dino/blob/de9ee3df6cf39fac952ab558447af1fa1365362a/vision_transformer.py#L174-L194, and
+        - https://github.com/facebookresearch/dinov2/blob/e1277af2ba9496fbadf7aec6eba56e8d882d1e35/dinov2/models/vision_transformer.py#L179-L211
+        """
+
+        num_patches = embeddings.shape[1]
+        num_positions = self.position_embedding.weight.shape[0]
+
+        # always interpolate when tracing to ensure the exported model works for dynamic input shapes
+        if not torch.jit.is_tracing() and num_patches == num_positions and height == width:
+            return self.position_embedding(self.position_ids)
+
+        patch_pos_embed = self.position_embedding.weight.unsqueeze(0)
+
+        dim = embeddings.shape[-1]
+
+        new_height = height // self.patch_size
+        new_width = width // self.patch_size
+
+        sqrt_num_positions = torch_int(num_positions**0.5)
+        patch_pos_embed = patch_pos_embed.reshape(1, sqrt_num_positions, sqrt_num_positions, dim)
+        patch_pos_embed = patch_pos_embed.permute(0, 3, 1, 2)
+
+        patch_pos_embed = nn.functional.interpolate(
+            patch_pos_embed,
+            size=(new_height, new_width),
+            mode="bicubic",
+            align_corners=False,
+        )
+
+        patch_pos_embed = patch_pos_embed.permute(0, 2, 3, 1).view(1, -1, dim)
+        return patch_pos_embed
+
+    def forward(self, pixel_values: torch.FloatTensor, interpolate_pos_encoding=False) -> torch.Tensor:
+        _, _, height, width = pixel_values.shape
+        target_dtype = self.patch_embedding.weight.dtype
+        patch_embeds = self.patch_embedding(pixel_values.to(dtype=target_dtype))  # shape = [*, width, grid, grid]
+        embeddings = patch_embeds.flatten(2).transpose(1, 2)
+
+        if interpolate_pos_encoding:
+            embeddings = embeddings + self.interpolate_pos_encoding(embeddings, height, width)
+        else:
+            embeddings = embeddings + self.position_embedding(self.position_ids)
+        return embeddings
+
+
+# Copied from transformers.models.clip.modeling_clip.CLIPTextEmbeddings with CLIP->Siglip
+class SiglipTextEmbeddings(nn.Module):
+    def __init__(self, config: SiglipTextConfig):
+        super().__init__()
+        embed_dim = config.hidden_size
+
+        self.token_embedding = nn.Embedding(config.vocab_size, embed_dim)
+        self.position_embedding = nn.Embedding(config.max_position_embeddings, embed_dim)
+
+        # position_ids (1, len position emb) is contiguous in memory and exported when serialized
+        self.register_buffer(
+            "position_ids", torch.arange(config.max_position_embeddings).expand((1, -1)), persistent=False
+        )
+
+    def forward(
+        self,
+        input_ids: Optional[torch.LongTensor] = None,
+        position_ids: Optional[torch.LongTensor] = None,
+        inputs_embeds: Optional[torch.FloatTensor] = None,
+    ) -> torch.Tensor:
+        seq_length = input_ids.shape[-1] if input_ids is not None else inputs_embeds.shape[-2]
+        max_position_embedding = self.position_embedding.weight.shape[0]
+
+        if seq_length > max_position_embedding:
+            raise ValueError(
+                f"Sequence length must be less than max_position_embeddings (got `sequence length`: "
+                f"{seq_length} and max_position_embeddings: {max_position_embedding}"
+            )
+
+        if position_ids is None:
+            position_ids = self.position_ids[:, :seq_length]
+
+        if inputs_embeds is None:
+            inputs_embeds = self.token_embedding(input_ids)
+
+        position_embeddings = self.position_embedding(position_ids)
+        embeddings = inputs_embeds + position_embeddings
+
+        return embeddings
+
+
+def eager_attention_forward(
+    module: nn.Module,
+    query: torch.Tensor,
+    key: torch.Tensor,
+    value: torch.Tensor,
+    attention_mask: Optional[torch.Tensor],
+    scaling: float,
+    dropout: float = 0.0,
+    **kwargs,
+):
+    attn_weights = torch.matmul(query, key.transpose(-1, -2)) * scaling
+    if attention_mask is not None:
+        attn_weights = attn_weights + attention_mask
+
+    attn_weights = nn.functional.softmax(attn_weights, dim=-1, dtype=torch.float32).to(query.dtype)
+    attn_weights = nn.functional.dropout(attn_weights, p=dropout, training=module.training)
+
+    attn_output = torch.matmul(attn_weights, value)
+    attn_output = attn_output.transpose(1, 2).contiguous()
+
+    return attn_output, attn_weights
+
+
+class SiglipAttention(nn.Module):
+    """Multi-headed attention from 'Attention Is All You Need' paper"""
+
+    def __init__(self, config):
+        super().__init__()
+        self.config = config
+        self.embed_dim = config.hidden_size
+        self.num_heads = config.num_attention_heads
+        self.head_dim = self.embed_dim // self.num_heads
+        if self.head_dim * self.num_heads != self.embed_dim:
+            raise ValueError(
+                f"embed_dim must be divisible by num_heads (got `embed_dim`: {self.embed_dim} and `num_heads`:"
+                f" {self.num_heads})."
+            )
+        self.scale = self.head_dim**-0.5
+        self.dropout = config.attention_dropout
+        self.is_causal = False
+
+        self.k_proj = nn.Linear(self.embed_dim, self.embed_dim)
+        self.v_proj = nn.Linear(self.embed_dim, self.embed_dim)
+        self.q_proj = nn.Linear(self.embed_dim, self.embed_dim)
+        self.out_proj = nn.Linear(self.embed_dim, self.embed_dim)
+
+    def forward(
+        self,
+        hidden_states: torch.Tensor,
+        attention_mask: Optional[torch.Tensor] = None,
+        output_attentions: Optional[bool] = False,
+    ) -> tuple[torch.Tensor, Optional[torch.Tensor]]:
+        """Input shape: Batch x Time x Channel"""
+
+        batch_size, seq_length, embed_dim = hidden_states.shape
+
+        queries = self.q_proj(hidden_states)
+        keys = self.k_proj(hidden_states)
+        values = self.v_proj(hidden_states)
+
+        queries = queries.view(batch_size, seq_length, self.num_heads, self.head_dim).transpose(1, 2)
+        keys = keys.view(batch_size, seq_length, self.num_heads, self.head_dim).transpose(1, 2)
+        values = values.view(batch_size, seq_length, self.num_heads, self.head_dim).transpose(1, 2)
+
+        attention_interface: Callable = eager_attention_forward
+        if self.config._attn_implementation != "eager":
+            if self.config._attn_implementation == "sdpa" and output_attentions:
+                logger.warning_once(
+                    "`torch.nn.functional.scaled_dot_product_attention` does not support `output_attentions=True`. Falling back to "
+                    'eager attention. This warning can be removed using the argument `attn_implementation="eager"` when loading the model.'
+                )
+            else:
+                attention_interface = ALL_ATTENTION_FUNCTIONS[self.config._attn_implementation]
+
+        attn_output, attn_weights = attention_interface(
+            self,
+            queries,
+            keys,
+            values,
+            attention_mask,
+            is_causal=self.is_causal,
+            scaling=self.scale,
+            dropout=0.0 if not self.training else self.dropout,
+        )
+
+        attn_output = attn_output.reshape(batch_size, seq_length, embed_dim).contiguous()
+        attn_output = self.out_proj(attn_output)
+
+        if not output_attentions:
+            attn_weights = None
+
+        return attn_output, attn_weights
+
+
+# Copied from transformers.models.clip.modeling_clip.CLIPMLP with CLIP->Siglip
+class SiglipMLP(nn.Module):
+    def __init__(self, config):
+        super().__init__()
+        self.config = config
+        self.activation_fn = ACT2FN[config.hidden_act]
+        self.fc1 = nn.Linear(config.hidden_size, config.intermediate_size)
+        self.fc2 = nn.Linear(config.intermediate_size, config.hidden_size)
+
+    def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
+        hidden_states = self.fc1(hidden_states)
+        hidden_states = self.activation_fn(hidden_states)
+        hidden_states = self.fc2(hidden_states)
+        return hidden_states
+
+
+class SiglipEncoderLayer(GradientCheckpointingLayer):
+    def __init__(self, config: Union[SiglipVisionConfig, SiglipTextConfig]):
+        super().__init__()
+        self.embed_dim = config.hidden_size
+        self.layer_norm1 = nn.LayerNorm(self.embed_dim, eps=config.layer_norm_eps)
+        self.self_attn = SiglipAttention(config)
+        self.layer_norm2 = nn.LayerNorm(self.embed_dim, eps=config.layer_norm_eps)
+        self.mlp = SiglipMLP(config)
+
+    def forward(
+        self,
+        hidden_states: torch.Tensor,
+        attention_mask: torch.Tensor,
+        output_attentions: Optional[bool] = False,
+    ) -> tuple[torch.FloatTensor]:
+        """
+        Args:
+            hidden_states (`torch.FloatTensor`):
+                Input to the layer of shape `(batch, seq_len, embed_dim)`.
+            attention_mask (`torch.FloatTensor`):
+                Attention mask of shape `(batch, 1, q_len, k_v_seq_len)` where padding elements are indicated by very large negative values.
+            output_attentions (`bool`, *optional*, defaults to `False`):
+                Whether or not to return the attentions tensors of all attention layers. See `attentions` under
+                returned tensors for more detail.
+        """
+        residual = hidden_states
+
+        hidden_states = self.layer_norm1(hidden_states)
+        hidden_states, attn_weights = self.self_attn(
+            hidden_states=hidden_states,
+            attention_mask=attention_mask,
+            output_attentions=output_attentions,
+        )
+        hidden_states = residual + hidden_states
+
+        residual = hidden_states
+        hidden_states = self.layer_norm2(hidden_states)
+        hidden_states = self.mlp(hidden_states)
+        hidden_states = residual + hidden_states
+
+        outputs = (hidden_states,)
+
+        if output_attentions:
+            outputs += (attn_weights,)
+
+        return outputs
+
+
+@auto_docstring
+class SiglipPreTrainedModel(PreTrainedModel):
+    config_class = SiglipConfig
+    base_model_prefix = "siglip"
+    supports_gradient_checkpointing = True
+
+    _no_split_modules = [
+        "SiglipTextEmbeddings",
+        "SiglipEncoderLayer",
+        "SiglipVisionEmbeddings",
+        "SiglipEncoderLayer",
+        "SiglipMultiheadAttentionPoolingHead",
+    ]
+    _supports_flash_attn_2 = True
+    _supports_sdpa = True
+    _supports_flex_attn = True
+    _supports_attention_backend = True
+
+    def _init_weights(self, module):
+        """Initialize the weights"""
+        if isinstance(module, SiglipVisionEmbeddings):
+            width = (
+                self.config.vision_config.hidden_size
+                if isinstance(self.config, SiglipConfig)
+                else self.config.hidden_size
+            )
+            nn.init.normal_(module.position_embedding.weight, std=1 / np.sqrt(width))
+        elif isinstance(module, nn.Embedding):
+            default_flax_embed_init(module.weight)
+        elif isinstance(module, SiglipAttention):
+            nn.init.xavier_uniform_(module.q_proj.weight)
+            nn.init.xavier_uniform_(module.k_proj.weight)
+            nn.init.xavier_uniform_(module.v_proj.weight)
+            nn.init.xavier_uniform_(module.out_proj.weight)
+            nn.init.zeros_(module.q_proj.bias)
+            nn.init.zeros_(module.k_proj.bias)
+            nn.init.zeros_(module.v_proj.bias)
+            nn.init.zeros_(module.out_proj.bias)
+        elif isinstance(module, SiglipMLP):
+            nn.init.xavier_uniform_(module.fc1.weight)
+            nn.init.xavier_uniform_(module.fc2.weight)
+            nn.init.normal_(module.fc1.bias, std=1e-6)
+            nn.init.normal_(module.fc2.bias, std=1e-6)
+        elif isinstance(module, SiglipMultiheadAttentionPoolingHead):
+            nn.init.xavier_uniform_(module.probe.data)
+            nn.init.xavier_uniform_(module.attention.in_proj_weight.data)
+            nn.init.zeros_(module.attention.in_proj_bias.data)
+        elif isinstance(module, SiglipModel):
+            logit_scale_init = torch.log(torch.tensor(1.0))
+            module.logit_scale.data.fill_(logit_scale_init)
+            module.logit_bias.data.zero_()
+        elif isinstance(module, SiglipForImageClassification):
+            nn.init.normal_(
+                module.classifier.weight,
+                std=self.config.vision_config.hidden_size**-0.5 * self.config.initializer_factor,
+            )
+        elif isinstance(module, (nn.Linear, nn.Conv2d)):
+            lecun_normal_(module.weight)
+            if module.bias is not None:
+                nn.init.zeros_(module.bias)
+        elif isinstance(module, nn.LayerNorm):
+            module.bias.data.zero_()
+            module.weight.data.fill_(1.0)
+
+
+# Copied from transformers.models.altclip.modeling_altclip.AltCLIPEncoder with AltCLIP->Siglip
+class SiglipEncoder(nn.Module):
+    """
+    Transformer encoder consisting of `config.num_hidden_layers` self attention layers. Each layer is a
+    [`SiglipEncoderLayer`].
+
+    Args:
+        config: SiglipConfig
+    """
+
+    def __init__(self, config: SiglipConfig):
+        super().__init__()
+        self.config = config
+        self.layers = nn.ModuleList([SiglipEncoderLayer(config) for _ in range(config.num_hidden_layers)])
+        self.gradient_checkpointing = False
+
+    # Ignore copy
+    @can_return_tuple
+    def forward(
+        self,
+        inputs_embeds,
+        attention_mask: Optional[torch.Tensor] = None,
+        output_attentions: Optional[bool] = None,
+        output_hidden_states: Optional[bool] = None,
+    ) -> BaseModelOutput:
+        r"""
+        Args:
+            inputs_embeds (`torch.FloatTensor` of shape `(batch_size, sequence_length, hidden_size)`):
+                Optionally, instead of passing `input_ids` you can choose to directly pass an embedded representation.
+                This is useful if you want more control over how to convert `input_ids` indices into associated vectors
+                than the model's internal embedding lookup matrix.
+            attention_mask (`torch.Tensor` of shape `(batch_size, sequence_length)`, *optional*):
+                Mask to avoid performing attention on padding token indices. Mask values selected in `[0, 1]`:
+
+                - 1 for tokens that are **not masked**,
+                - 0 for tokens that are **masked**.
+
+                [What are attention masks?](../glossary#attention-mask)
+            output_attentions (`bool`, *optional*):
+                Whether or not to return the attentions tensors of all attention layers. See `attentions` under
+                returned tensors for more detail.
+            output_hidden_states (`bool`, *optional*):
+                Whether or not to return the hidden states of all layers. See `hidden_states` under returned tensors
+                for more detail.
+            return_dict (`bool`, *optional*):
+                Whether or not to return a [`~utils.ModelOutput`] instead of a plain tuple.
+        """
+        output_attentions = output_attentions if output_attentions is not None else self.config.output_attentions
+        output_hidden_states = (
+            output_hidden_states if output_hidden_states is not None else self.config.output_hidden_states
+        )
+
+        encoder_states = () if output_hidden_states else None
+        all_attentions = () if output_attentions else None
+
+        hidden_states = inputs_embeds
+        for encoder_layer in self.layers:
+            if output_hidden_states:
+                encoder_states = encoder_states + (hidden_states,)
+
+            layer_outputs = encoder_layer(
+                hidden_states,
+                attention_mask,
+                output_attentions=output_attentions,
+            )
+
+            hidden_states = layer_outputs[0]
+
+            if output_attentions:
+                all_attentions = all_attentions + (layer_outputs[1],)
+
+        if output_hidden_states:
+            encoder_states = encoder_states + (hidden_states,)
+
+        return BaseModelOutput(
+            last_hidden_state=hidden_states,
+            hidden_states=encoder_states,
+            attentions=all_attentions,
+        )
+
+
+class SiglipTextTransformer(nn.Module):
+    def __init__(self, config: SiglipTextConfig):
+        super().__init__()
+        self.config = config
+        embed_dim = config.hidden_size
+        self.embeddings = SiglipTextEmbeddings(config)
+        self.encoder = SiglipEncoder(config)
+        self.final_layer_norm = nn.LayerNorm(embed_dim, eps=config.layer_norm_eps)
+
+        self.head = nn.Linear(embed_dim, config.projection_size)
+        self._use_flash_attention_2 = config._attn_implementation == "flash_attention_2"
+
+    @can_return_tuple
+    @auto_docstring
+    def forward(
+        self,
+        input_ids: Optional[torch.Tensor] = None,
+        attention_mask: Optional[torch.Tensor] = None,
+        position_ids: Optional[torch.Tensor] = None,
+        output_attentions: Optional[bool] = None,
+        output_hidden_states: Optional[bool] = None,
+    ) -> BaseModelOutputWithPooling:
+        output_attentions = output_attentions if output_attentions is not None else self.config.output_attentions
+        output_hidden_states = (
+            output_hidden_states if output_hidden_states is not None else self.config.output_hidden_states
+        )
+
+        if input_ids is None:
+            raise ValueError("You have to specify input_ids")
+
+        input_shape = input_ids.size()
+        input_ids = input_ids.view(-1, input_shape[-1])
+
+        hidden_states = self.embeddings(input_ids=input_ids, position_ids=position_ids)
+
+        # note: SigLIP's text model does not use a causal mask, unlike the original CLIP model.
+        # expand attention_mask
+        if attention_mask is not None and not self._use_flash_attention_2:
+            # [batch_size, seq_len] -> [batch_size, 1, tgt_seq_len, src_seq_len]
+            attention_mask = _prepare_4d_attention_mask(attention_mask, hidden_states.dtype)
+
+        encoder_outputs: BaseModelOutput = self.encoder(
+            inputs_embeds=hidden_states,
+            attention_mask=attention_mask,
+            output_attentions=output_attentions,
+            output_hidden_states=output_hidden_states,
+        )
+
+        last_hidden_state = encoder_outputs.last_hidden_state
+        last_hidden_state = self.final_layer_norm(last_hidden_state)
+
+        # Assuming "sticky" EOS tokenization, last token is always EOS.
+        pooled_output = last_hidden_state[:, -1, :]
+        pooled_output = self.head(pooled_output)
+
+        return BaseModelOutputWithPooling(
+            last_hidden_state=last_hidden_state,
+            pooler_output=pooled_output,
+            hidden_states=encoder_outputs.hidden_states,
+            attentions=encoder_outputs.attentions,
+        )
+
+
+@auto_docstring(
+    custom_intro="""
+    The text model from SigLIP without any head or projection on top.
+    """
+)
+class SiglipTextModel(SiglipPreTrainedModel):
+    config_class = SiglipTextConfig
+
+    def __init__(self, config: SiglipTextConfig):
+        super().__init__(config)
+        self.text_model = SiglipTextTransformer(config)
+        # Initialize weights and apply final processing
+        self.post_init()
+
+    def get_input_embeddings(self) -> nn.Module:
+        return self.text_model.embeddings.token_embedding
+
+    def set_input_embeddings(self, value):
+        self.text_model.embeddings.token_embedding = value
+
+    @can_return_tuple
+    @auto_docstring
+    def forward(
+        self,
+        input_ids: Optional[torch.Tensor] = None,
+        attention_mask: Optional[torch.Tensor] = None,
+        position_ids: Optional[torch.Tensor] = None,
+        output_attentions: Optional[bool] = None,
+        output_hidden_states: Optional[bool] = None,
+    ) -> BaseModelOutputWithPooling:
+        r"""
+        Examples:
+
+        ```python
+        >>> from transformers import AutoTokenizer, SiglipTextModel
+
+        >>> model = SiglipTextModel.from_pretrained("google/siglip-base-patch16-224")
+        >>> tokenizer = AutoTokenizer.from_pretrained("google/siglip-base-patch16-224")
+
+        >>> # important: make sure to set padding="max_length" as that's how the model was trained
+        >>> inputs = tokenizer(["a photo of a cat", "a photo of a dog"], padding="max_length", return_tensors="pt")
+
+        >>> outputs = model(**inputs)
+        >>> last_hidden_state = outputs.last_hidden_state
+        >>> pooled_output = outputs.pooler_output  # pooled (EOS token) states
+        ```"""
+
+        return self.text_model(
+            input_ids=input_ids,
+            attention_mask=attention_mask,
+            position_ids=position_ids,
+            output_attentions=output_attentions,
+            output_hidden_states=output_hidden_states,
+        )
+
+
+class SiglipVisionTransformer(nn.Module):
+    def __init__(self, config: SiglipVisionConfig):
+        super().__init__()
+        self.config = config
+        embed_dim = config.hidden_size
+
+        self.embeddings = SiglipVisionEmbeddings(config)
+        self.encoder = SiglipEncoder(config)
+        self.post_layernorm = nn.LayerNorm(embed_dim, eps=config.layer_norm_eps)
+        self.use_head = True if not hasattr(config, "vision_use_head") else config.vision_use_head
+        if self.use_head:
+            self.head = SiglipMultiheadAttentionPoolingHead(config)
+
+    @can_return_tuple
+    @auto_docstring
+    def forward(
+        self,
+        pixel_values,
+        output_attentions: Optional[bool] = None,
+        output_hidden_states: Optional[bool] = None,
+        interpolate_pos_encoding: Optional[bool] = False,
+    ) -> BaseModelOutputWithPooling:
+        output_attentions = output_attentions if output_attentions is not None else self.config.output_attentions
+        output_hidden_states = (
+            output_hidden_states if output_hidden_states is not None else self.config.output_hidden_states
+        )
+
+        hidden_states = self.embeddings(pixel_values, interpolate_pos_encoding=interpolate_pos_encoding)
+        # Convert to bfloat16 if the encoder uses bfloat16
+        if len(self.encoder.layers) > 0 and self.encoder.layers[0].self_attn.q_proj.weight.dtype == torch.bfloat16:
+            hidden_states = hidden_states.to(torch.bfloat16)
+
+        encoder_outputs: BaseModelOutput = self.encoder(
+            inputs_embeds=hidden_states,
+            output_attentions=output_attentions,
+            output_hidden_states=output_hidden_states,
+        )
+
+        last_hidden_state = encoder_outputs.last_hidden_state
+        last_hidden_state = self.post_layernorm(last_hidden_state)
+
+        pooler_output = self.head(last_hidden_state) if self.use_head else None
+
+        return BaseModelOutputWithPooling(
+            last_hidden_state=last_hidden_state,
+            pooler_output=pooler_output,
+            hidden_states=encoder_outputs.hidden_states,
+            attentions=encoder_outputs.attentions,
+        )
+
+
+class SiglipMultiheadAttentionPoolingHead(nn.Module):
+    """Multihead Attention Pooling."""
+
+    def __init__(self, config: SiglipVisionConfig):
+        super().__init__()
+
+        self.probe = nn.Parameter(torch.randn(1, 1, config.hidden_size))
+        self.attention = torch.nn.MultiheadAttention(config.hidden_size, config.num_attention_heads, batch_first=True)
+        self.layernorm = nn.LayerNorm(config.hidden_size, eps=config.layer_norm_eps)
+        self.mlp = SiglipMLP(config)
+
+    def forward(self, hidden_state):
+        batch_size = hidden_state.shape[0]
+        probe = self.probe.repeat(batch_size, 1, 1)
+
+        hidden_state = self.attention(probe, hidden_state, hidden_state)[0]
+
+        residual = hidden_state
+        hidden_state = self.layernorm(hidden_state)
+        hidden_state = residual + self.mlp(hidden_state)
+
+        return hidden_state[:, 0]
+
+
+@auto_docstring(
+    custom_intro="""
+    The vision model from SigLIP without any head or projection on top.
+    """
+)
+class SiglipVisionModel(SiglipPreTrainedModel):
+    config_class = SiglipVisionConfig
+    main_input_name = "pixel_values"
+
+    def __init__(self, config: SiglipVisionConfig):
+        super().__init__(config)
+
+        self.vision_model = SiglipVisionTransformer(config)
+
+        # Initialize weights and apply final processing
+        self.post_init()
+
+    def get_input_embeddings(self) -> nn.Module:
+        return self.vision_model.embeddings.patch_embedding
+
+    @can_return_tuple
+    @auto_docstring
+    def forward(
+        self,
+        pixel_values,
+        output_attentions: Optional[bool] = None,
+        output_hidden_states: Optional[bool] = None,
+        interpolate_pos_encoding: bool = False,
+    ) -> BaseModelOutputWithPooling:
+        r"""
+        Examples:
+
+        ```python
+        >>> from PIL import Image
+        >>> import requests
+        >>> from transformers import AutoProcessor, SiglipVisionModel
+
+        >>> model = SiglipVisionModel.from_pretrained("google/siglip-base-patch16-224")
+        >>> processor = AutoProcessor.from_pretrained("google/siglip-base-patch16-224")
+
+        >>> url = "http://images.cocodataset.org/val2017/000000039769.jpg"
+        >>> image = Image.open(requests.get(url, stream=True).raw)
+
+        >>> inputs = processor(images=image, return_tensors="pt")
+
+        >>> outputs = model(**inputs)
+        >>> last_hidden_state = outputs.last_hidden_state
+        >>> pooled_output = outputs.pooler_output  # pooled features
+        ```"""
+
+        return self.vision_model(
+            pixel_values=pixel_values,
+            output_attentions=output_attentions,
+            output_hidden_states=output_hidden_states,
+            interpolate_pos_encoding=interpolate_pos_encoding,
+        )
+
+
+@auto_docstring
+class SiglipModel(SiglipPreTrainedModel):
+    config_class = SiglipConfig
+
+    def __init__(self, config: SiglipConfig):
+        super().__init__(config)
+
+        if not isinstance(config.text_config, SiglipTextConfig):
+            raise TypeError(
+                "config.text_config is expected to be of type SiglipTextConfig but is of type"
+                f" {type(config.text_config)}."
+            )
+
+        if not isinstance(config.vision_config, SiglipVisionConfig):
+            raise TypeError(
+                "config.vision_config is expected to be of type SiglipVisionConfig but is of type"
+                f" {type(config.vision_config)}."
+            )
+
+        text_config = config.text_config
+        vision_config = config.vision_config
+
+        # First, initialize the text and vision models with proper attention implementation
+        text_model = SiglipTextModel._from_config(text_config)
+        vision_model = SiglipVisionModel._from_config(vision_config)
+
+        # Second, get the text and vision submodules (for backward compatibility)
+        self.text_model = text_model.text_model
+        self.vision_model = vision_model.vision_model
+
+        self.logit_scale = nn.Parameter(torch.randn(1))
+        self.logit_bias = nn.Parameter(torch.randn(1))
+
+        # Initialize weights and apply final processing
+        self.post_init()
+
+    @auto_docstring
+    def get_text_features(
+        self,
+        input_ids: Optional[torch.Tensor] = None,
+        attention_mask: Optional[torch.Tensor] = None,
+        position_ids: Optional[torch.Tensor] = None,
+        output_attentions: Optional[bool] = None,
+        output_hidden_states: Optional[bool] = None,
+    ) -> torch.FloatTensor:
+        r"""
+        Returns:
+            text_features (`torch.FloatTensor` of shape `(batch_size, output_dim`): The text embeddings obtained by
+            applying the projection layer to the pooled output of [`SiglipTextModel`].
+
+        Examples:
+
+        ```python
+        >>> from transformers import AutoTokenizer, AutoModel
+        >>> import torch
+
+        >>> model = AutoModel.from_pretrained("google/siglip-base-patch16-224")
+        >>> tokenizer = AutoTokenizer.from_pretrained("google/siglip-base-patch16-224")
+
+        >>> # important: make sure to set padding="max_length" as that's how the model was trained
+        >>> inputs = tokenizer(["a photo of a cat", "a photo of a dog"], padding="max_length", return_tensors="pt")
+        >>> with torch.no_grad():
+        ...     text_features = model.get_text_features(**inputs)
+        ```"""
+        # Use SigLIP model's config for some fields (if specified) instead of those of vision & text components.
+        output_attentions = output_attentions if output_attentions is not None else self.config.output_attentions
+        output_hidden_states = (
+            output_hidden_states if output_hidden_states is not None else self.config.output_hidden_states
+        )
+
+        text_outputs: BaseModelOutputWithPooling = self.text_model(
+            input_ids=input_ids,
+            attention_mask=attention_mask,
+            position_ids=position_ids,
+            output_attentions=output_attentions,
+            output_hidden_states=output_hidden_states,
+        )
+
+        pooled_output = text_outputs.pooler_output
+
+        return pooled_output
+
+    @auto_docstring
+    def get_image_features(
+        self,
+        pixel_values: Optional[torch.FloatTensor] = None,
+        output_attentions: Optional[bool] = None,
+        output_hidden_states: Optional[bool] = None,
+        interpolate_pos_encoding: bool = False,
+    ) -> torch.FloatTensor:
+        r"""
+        Returns:
+            image_features (`torch.FloatTensor` of shape `(batch_size, output_dim`): The image embeddings obtained by
+            applying the projection layer to the pooled output of [`SiglipVisionModel`].
+
+        Examples:
+
+        ```python
+        >>> from PIL import Image
+        >>> import requests
+        >>> from transformers import AutoProcessor, AutoModel
+        >>> import torch
+
+        >>> model = AutoModel.from_pretrained("google/siglip-base-patch16-224")
+        >>> processor = AutoProcessor.from_pretrained("google/siglip-base-patch16-224")
+
+        >>> url = "http://images.cocodataset.org/val2017/000000039769.jpg"
+        >>> image = Image.open(requests.get(url, stream=True).raw)
+
+        >>> inputs = processor(images=image, return_tensors="pt")
+
+        >>> with torch.no_grad():
+        ...     image_features = model.get_image_features(**inputs)
+        ```"""
+        # Use SiglipModel's config for some fields (if specified) instead of those of vision & text components.
+        output_attentions = output_attentions if output_attentions is not None else self.config.output_attentions
+        output_hidden_states = (
+            output_hidden_states if output_hidden_states is not None else self.config.output_hidden_states
+        )
+
+        vision_outputs: BaseModelOutputWithPooling = self.vision_model(
+            pixel_values=pixel_values,
+            output_attentions=output_attentions,
+            output_hidden_states=output_hidden_states,
+            interpolate_pos_encoding=interpolate_pos_encoding,
+        )
+
+        pooled_output = vision_outputs.pooler_output
+
+        return pooled_output
+
+    @can_return_tuple
+    @auto_docstring
+    def forward(
+        self,
+        input_ids: Optional[torch.LongTensor] = None,
+        pixel_values: Optional[torch.FloatTensor] = None,
+        attention_mask: Optional[torch.Tensor] = None,
+        position_ids: Optional[torch.LongTensor] = None,
+        return_loss: Optional[bool] = None,
+        output_attentions: Optional[bool] = None,
+        output_hidden_states: Optional[bool] = None,
+        interpolate_pos_encoding: bool = False,
+    ) -> SiglipOutput:
+        r"""
+        return_loss (`bool`, *optional*):
+            Whether or not to return the contrastive loss.
+
+        Examples:
+
+        ```python
+        >>> from PIL import Image
+        >>> import requests
+        >>> from transformers import AutoProcessor, AutoModel
+        >>> import torch
+
+        >>> model = AutoModel.from_pretrained("google/siglip-base-patch16-224")
+        >>> processor = AutoProcessor.from_pretrained("google/siglip-base-patch16-224")
+
+        >>> url = "http://images.cocodataset.org/val2017/000000039769.jpg"
+        >>> image = Image.open(requests.get(url, stream=True).raw)
+
+        >>> texts = ["a photo of 2 cats", "a photo of 2 dogs"]
+        >>> # important: we pass `padding=max_length` since the model was trained with this
+        >>> inputs = processor(text=texts, images=image, padding="max_length", return_tensors="pt")
+
+        >>> with torch.no_grad():
+        ...     outputs = model(**inputs)
+
+        >>> logits_per_image = outputs.logits_per_image
+        >>> probs = torch.sigmoid(logits_per_image) # these are the probabilities
+        >>> print(f"{probs[0][0]:.1%} that image 0 is '{texts[0]}'")
+        31.9% that image 0 is 'a photo of 2 cats'
+        ```"""
+        # Use SigLIP model's config for some fields (if specified) instead of those of vision & text components.
+        output_attentions = output_attentions if output_attentions is not None else self.config.output_attentions
+        output_hidden_states = (
+            output_hidden_states if output_hidden_states is not None else self.config.output_hidden_states
+        )
+
+        vision_outputs: BaseModelOutputWithPooling = self.vision_model(
+            pixel_values=pixel_values,
+            output_attentions=output_attentions,
+            output_hidden_states=output_hidden_states,
+            interpolate_pos_encoding=interpolate_pos_encoding,
+        )
+
+        text_outputs: BaseModelOutputWithPooling = self.text_model(
+            input_ids=input_ids,
+            attention_mask=attention_mask,
+            position_ids=position_ids,
+            output_attentions=output_attentions,
+            output_hidden_states=output_hidden_states,
+        )
+
+        image_embeds = vision_outputs.pooler_output
+        text_embeds = text_outputs.pooler_output
+
+        # normalized features
+        image_embeds = image_embeds / image_embeds.norm(p=2, dim=-1, keepdim=True)
+        text_embeds = text_embeds / text_embeds.norm(p=2, dim=-1, keepdim=True)
+
+        # cosine similarity as logits
+        logits_per_text = torch.matmul(text_embeds, image_embeds.t().to(text_embeds.device))
+
+        logit_scale, logit_bias = self.logit_scale.to(text_embeds.device), self.logit_bias.to(text_embeds.device)
+        logits_per_text = logits_per_text * logit_scale.exp() + logit_bias
+
+        logits_per_image = logits_per_text.t()
+
+        loss = None
+        if return_loss:
+            # Adapted from https://github.com/google-research/big_vision/blob/01edb81a4716f93a48be43b3a4af14e29cdb3a7f/big_vision/trainers/proj/image_text/siglip.py#L287
+            eye = torch.eye(logits_per_text.size(0), device=logits_per_text.device)
+            m1_diag1 = -torch.ones_like(logits_per_text) + 2 * eye
+            loglik = torch.nn.functional.logsigmoid(m1_diag1 * logits_per_text)
+            nll = -torch.sum(loglik, dim=-1)
+            loss = nll.mean()
+
+        return SiglipOutput(
+            loss=loss,
+            logits_per_image=logits_per_image,
+            logits_per_text=logits_per_text,
+            text_embeds=text_embeds,
+            image_embeds=image_embeds,
+            text_model_output=text_outputs,
+            vision_model_output=vision_outputs,
+        )
+
+
+@auto_docstring(
+    custom_intro="""
+    SigLIP vision encoder with an image classification head on top (a linear layer on top of the pooled final hidden states of
+    the patch tokens) e.g. for ImageNet.
+    """
+)
+class SiglipForImageClassification(SiglipPreTrainedModel):
+    main_input_name = "pixel_values"
+
+    def __init__(self, config: SiglipConfig) -> None:
+        super().__init__(config)
+
+        self.num_labels = config.num_labels
+
+        # Create the vision model with proper attention
+        # and take only vision_model submodule (for backward compatibility)
+        vision_model = SiglipVisionModel._from_config(config.vision_config)
+        self.vision_model = vision_model.vision_model
+
+        # Classifier head
+        self.classifier = (
+            nn.Linear(config.vision_config.hidden_size, config.num_labels) if config.num_labels > 0 else nn.Identity()
+        )
+
+        # Initialize weights and apply final processing
+        self.post_init()
+
+    @can_return_tuple
+    @auto_docstring
+    def forward(
+        self,
+        pixel_values: Optional[torch.Tensor] = None,
+        labels: Optional[torch.Tensor] = None,
+        output_attentions: Optional[bool] = None,
+        output_hidden_states: Optional[bool] = None,
+        interpolate_pos_encoding: bool = False,
+    ) -> ImageClassifierOutput:
+        r"""
+        labels (`torch.LongTensor` of shape `(batch_size,)`, *optional*):
+            Labels for computing the image classification/regression loss. Indices should be in `[0, ...,
+            config.num_labels - 1]`. If `config.num_labels == 1` a regression loss is computed (Mean-Square loss), If
+            `config.num_labels > 1` a classification loss is computed (Cross-Entropy).
+
+        Examples:
+
+        ```python
+        >>> from transformers import AutoImageProcessor, SiglipForImageClassification
+        >>> import torch
+        >>> from PIL import Image
+        >>> import requests
+
+        >>> torch.manual_seed(3)  # doctest: +IGNORE_RESULT
+        >>> url = "http://images.cocodataset.org/val2017/000000039769.jpg"
+        >>> image = Image.open(requests.get(url, stream=True).raw)
+
+        >>> # note: we are loading a `SiglipModel` from the hub here,
+        >>> # so the head will be randomly initialized, hence the predictions will be random if seed is not set above.
+        >>> image_processor = AutoImageProcessor.from_pretrained("google/siglip-base-patch16-224")
+        >>> model = SiglipForImageClassification.from_pretrained("google/siglip-base-patch16-224")
+
+        >>> inputs = image_processor(images=image, return_tensors="pt")
+        >>> outputs = model(**inputs)
+        >>> logits = outputs.logits
+        >>> # model predicts one of the two classes
+        >>> predicted_class_idx = logits.argmax(-1).item()
+        >>> print("Predicted class:", model.config.id2label[predicted_class_idx])
+        Predicted class: LABEL_1
+        ```"""
+        output_attentions = output_attentions if output_attentions is not None else self.config.output_attentions
+        output_hidden_states = (
+            output_hidden_states if output_hidden_states is not None else self.config.output_hidden_states
+        )
+
+        outputs: BaseModelOutputWithPooling = self.vision_model(
+            pixel_values,
+            output_attentions=output_attentions,
+            output_hidden_states=output_hidden_states,
+            interpolate_pos_encoding=interpolate_pos_encoding,
+        )
+
+        sequence_output = outputs.last_hidden_state
+
+        # average pool the patch tokens
+        sequence_output = torch.mean(sequence_output, dim=1)
+        # apply classifier
+        logits = self.classifier(sequence_output)
+
+        loss = None
+        if labels is not None:
+            # move labels to correct device to enable model parallelism
+            labels = labels.to(logits.device)
+            if self.config.problem_type is None:
+                if self.num_labels == 1:
+                    self.config.problem_type = "regression"
+                elif self.num_labels > 1 and (labels.dtype == torch.long or labels.dtype == torch.int):
+                    self.config.problem_type = "single_label_classification"
+                else:
+                    self.config.problem_type = "multi_label_classification"
+
+            if self.config.problem_type == "regression":
+                loss_fct = MSELoss()
+                if self.num_labels == 1:
+                    loss = loss_fct(logits.squeeze(), labels.squeeze())
+                else:
+                    loss = loss_fct(logits, labels)
+            elif self.config.problem_type == "single_label_classification":
+                loss_fct = CrossEntropyLoss()
+                loss = loss_fct(logits.view(-1, self.num_labels), labels.view(-1))
+            elif self.config.problem_type == "multi_label_classification":
+                loss_fct = BCEWithLogitsLoss()
+                loss = loss_fct(logits, labels)
+
+        return ImageClassifierOutput(
+            loss=loss,
+            logits=logits,
+            hidden_states=outputs.hidden_states,
+            attentions=outputs.attentions,
+        )
+
+
+__all__ = [
+    "SiglipModel",
+    "SiglipPreTrainedModel",
+    "SiglipTextModel",
+    "SiglipVisionModel",
+    "SiglipForImageClassification",
+]

real_inference/source/XPolicyLab/policy/Pi_05/openpi/src/openpi/policies/aloha_policy.py

@@ -0,0 +1,209 @@
+import dataclasses
+from typing import ClassVar
+
+import einops
+import numpy as np
+
+from openpi import transforms
+
+
+def make_aloha_example() -> dict:
+    """Creates a random input example for the Aloha policy."""
+    return {
+        "state": np.ones((14,)),
+        "images": {
+            "cam_high": np.random.randint(256, size=(3, 224, 224), dtype=np.uint8),
+            "cam_low": np.random.randint(256, size=(3, 224, 224), dtype=np.uint8),
+            "cam_left_wrist": np.random.randint(256, size=(3, 224, 224), dtype=np.uint8),
+            "cam_right_wrist": np.random.randint(256, size=(3, 224, 224), dtype=np.uint8),
+        },
+        "prompt": "do something",
+    }
+
+
+def _make_image_mask(image: np.ndarray, valid: bool) -> np.ndarray | np.bool_:
+    if image.ndim == 4:
+        return np.full((image.shape[0],), valid, dtype=bool)
+    return np.bool_(valid)
+
+
+@dataclasses.dataclass(frozen=True)
+class AlohaInputs(transforms.DataTransformFn):
+    """Inputs for the Aloha policy.
+
+    Expected inputs:
+    - images: dict[name, img] where img is [channel, height, width]. name must be in EXPECTED_CAMERAS.
+    - state: [14]
+    - actions: [action_horizon, 14]
+    """
+
+    # If true, this will convert the joint and gripper values from the standard Aloha space to
+    # the space used by the pi internal runtime which was used to train the base model.
+    adapt_to_pi: bool = True
+
+    # The expected cameras names. All input cameras must be in this set. Missing cameras will be
+    # replaced with black images and the corresponding `image_mask` will be set to False.
+    EXPECTED_CAMERAS: ClassVar[tuple[str, ...]] = ("cam_high", "cam_low", "cam_left_wrist", "cam_right_wrist")
+
+    def __call__(self, data: dict) -> dict:
+        data = _decode_aloha(data, adapt_to_pi=self.adapt_to_pi)
+
+        in_images = data["images"]
+        if set(in_images) - set(self.EXPECTED_CAMERAS):
+            raise ValueError(f"Expected images to contain {self.EXPECTED_CAMERAS}, got {tuple(in_images)}")
+
+        # Assume that base image always exists.
+        base_image = in_images["cam_high"]
+
+        images = {
+            "base_0_rgb": base_image,
+        }
+        image_masks = {
+            "base_0_rgb": _make_image_mask(base_image, True),
+        }
+
+        # Add the extra images.
+        extra_image_names = {
+            "left_wrist_0_rgb": "cam_left_wrist",
+            "right_wrist_0_rgb": "cam_right_wrist",
+        }
+        for dest, source in extra_image_names.items():
+            if source in in_images:
+                images[dest] = in_images[source]
+                image_masks[dest] = _make_image_mask(in_images[source], True)
+            else:
+                images[dest] = np.zeros_like(base_image)
+                image_masks[dest] = _make_image_mask(base_image, False)
+
+        inputs = {
+            "image": images,
+            "image_mask": image_masks,
+            "state": data["state"],
+        }
+
+        # Actions are only available during training.
+        if "actions" in data:
+            actions = np.asarray(data["actions"])
+            actions = _encode_actions_inv(actions, adapt_to_pi=self.adapt_to_pi)
+            inputs["actions"] = actions
+
+        if "prompt" in data:
+            inputs["prompt"] = data["prompt"]
+
+        return inputs
+
+
+@dataclasses.dataclass(frozen=True)
+class AlohaOutputs(transforms.DataTransformFn):
+    """Outputs for the Aloha policy."""
+
+    # If true, this will convert the joint and gripper values from the standard Aloha space to
+    # the space used by the pi internal runtime which was used to train the base model.
+    adapt_to_pi: bool = True
+
+    def __call__(self, data: dict) -> dict:
+        # Only return the first 14 dims.
+        actions = np.asarray(data["actions"])[..., :14]
+        return {"actions": _encode_actions(actions, adapt_to_pi=self.adapt_to_pi)}
+
+
+def _joint_flip_mask() -> np.ndarray:
+    """Used to convert between aloha and pi joint angles."""
+    return np.array([1, -1, -1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1])
+
+
+def _normalize(x, min_val, max_val):
+    return (x - min_val) / (max_val - min_val)
+
+
+def _unnormalize(x, min_val, max_val):
+    return x * (max_val - min_val) + min_val
+
+
+def _gripper_to_angular(value):
+    # Aloha transforms the gripper positions into a linear space. The following code
+    # reverses this transformation to be consistent with pi0 which is pretrained in
+    # angular space.
+    #
+    # These values are coming from the Aloha code:
+    # PUPPET_GRIPPER_POSITION_OPEN, PUPPET_GRIPPER_POSITION_CLOSED
+    value = _unnormalize(value, min_val=0.01844, max_val=0.05800)
+
+    # This is the inverse of the angular to linear transformation inside the Interbotix code.
+    def linear_to_radian(linear_position, arm_length, horn_radius):
+        value = (horn_radius**2 + linear_position**2 - arm_length**2) / (2 * horn_radius * linear_position)
+        return np.arcsin(np.clip(value, -1.0, 1.0))
+
+    # The constants are taken from the Interbotix code.
+    value = linear_to_radian(value, arm_length=0.036, horn_radius=0.022)
+
+    # pi0 gripper data is normalized (0, 1) between encoder counts (2405, 3110).
+    # There are 4096 total encoder counts and aloha uses a zero of 2048.
+    # Converting this to radians means that the normalized inputs are between (0.5476, 1.6296)
+    return _normalize(value, min_val=0.5476, max_val=1.6296)
+
+
+def _gripper_from_angular(value):
+    # Convert from the gripper position used by pi0 to the gripper position that is used by Aloha.
+    # Note that the units are still angular but the range is different.
+
+    # We do not scale the output since the trossen model predictions are already in radians.
+    # See the comment in _gripper_to_angular for a derivation of the constant
+    value = value + 0.5476
+
+    # These values are coming from the Aloha code:
+    # PUPPET_GRIPPER_JOINT_OPEN, PUPPET_GRIPPER_JOINT_CLOSE
+    return _normalize(value, min_val=-0.6213, max_val=1.4910)
+
+
+def _gripper_from_angular_inv(value):
+    # Directly inverts the gripper_from_angular function.
+    value = _unnormalize(value, min_val=-0.6213, max_val=1.4910)
+    return value - 0.5476
+
+
+def _decode_aloha(data: dict, *, adapt_to_pi: bool = False) -> dict:
+    # state is [left_arm_joint_angles, left_arm_gripper, right_arm_joint_angles, right_arm_gripper]
+    # dim sizes: [6, 1, 6, 1]
+    state = np.asarray(data["state"])
+    state = _decode_state(state, adapt_to_pi=adapt_to_pi)
+
+    def convert_image(img):
+        img = np.asarray(img)
+        # Convert to uint8 if using float images.
+        if np.issubdtype(img.dtype, np.floating):
+            img = (255 * img).astype(np.uint8)
+        if img.ndim == 4:
+            return einops.rearrange(img, "b c h w -> b h w c")
+        return einops.rearrange(img, "c h w -> h w c")
+
+    images = data["images"]
+    images_dict = {name: convert_image(img) for name, img in images.items()}
+
+    data["images"] = images_dict
+    data["state"] = state
+    return data
+
+
+def _decode_state(state: np.ndarray, *, adapt_to_pi: bool = False) -> np.ndarray:
+    if adapt_to_pi:
+        # Flip the joints.
+        state = _joint_flip_mask() * state
+        # Reverse the gripper transformation that is being applied by the Aloha runtime.
+        state[..., [6, 13]] = _gripper_to_angular(state[..., [6, 13]])
+    return state
+
+
+def _encode_actions(actions: np.ndarray, *, adapt_to_pi: bool = False) -> np.ndarray:
+    if adapt_to_pi:
+        # Flip the joints.
+        actions = _joint_flip_mask() * actions
+        actions[..., [6, 13]] = _gripper_from_angular(actions[..., [6, 13]])
+    return actions
+
+
+def _encode_actions_inv(actions: np.ndarray, *, adapt_to_pi: bool = False) -> np.ndarray:
+    if adapt_to_pi:
+        actions = _joint_flip_mask() * actions
+        actions[..., [6, 13]] = _gripper_from_angular_inv(actions[..., [6, 13]])
+    return actions

real_inference/source/XPolicyLab/policy/Pi_05/openpi/src/openpi/policies/droid_policy.py

@@ -0,0 +1,81 @@
+import dataclasses
+
+import einops
+import numpy as np
+
+from openpi import transforms
+from openpi.models import model as _model
+
+
+def make_droid_example() -> dict:
+    """Creates a random input example for the Droid policy."""
+    return {
+        "observation/exterior_image_1_left": np.random.randint(256, size=(224, 224, 3), dtype=np.uint8),
+        "observation/wrist_image_left": np.random.randint(256, size=(224, 224, 3), dtype=np.uint8),
+        "observation/joint_position": np.random.rand(7),
+        "observation/gripper_position": np.random.rand(1),
+        "prompt": "do something",
+    }
+
+
+def _parse_image(image) -> np.ndarray:
+    image = np.asarray(image)
+    if np.issubdtype(image.dtype, np.floating):
+        image = (255 * image).astype(np.uint8)
+    if image.shape[0] == 3:
+        image = einops.rearrange(image, "c h w -> h w c")
+    return image
+
+
+@dataclasses.dataclass(frozen=True)
+class DroidInputs(transforms.DataTransformFn):
+    # Determines which model will be used.
+    model_type: _model.ModelType
+
+    def __call__(self, data: dict) -> dict:
+        gripper_pos = np.asarray(data["observation/gripper_position"])
+        if gripper_pos.ndim == 0:
+            # Ensure gripper position is a 1D array, not a scalar, so we can concatenate with joint positions
+            gripper_pos = gripper_pos[np.newaxis]
+        state = np.concatenate([data["observation/joint_position"], gripper_pos])
+
+        # Possibly need to parse images to uint8 (H,W,C) since LeRobot automatically
+        # stores as float32 (C,H,W), gets skipped for policy inference
+        base_image = _parse_image(data["observation/exterior_image_1_left"])
+        wrist_image = _parse_image(data["observation/wrist_image_left"])
+
+        match self.model_type:
+            case _model.ModelType.PI0 | _model.ModelType.PI05:
+                names = ("base_0_rgb", "left_wrist_0_rgb", "right_wrist_0_rgb")
+                images = (base_image, wrist_image, np.zeros_like(base_image))
+                image_masks = (np.True_, np.True_, np.False_)
+            case _model.ModelType.PI0_FAST:
+                names = ("base_0_rgb", "base_1_rgb", "wrist_0_rgb")
+                # We don't mask out padding images for FAST models.
+                images = (base_image, np.zeros_like(base_image), wrist_image)
+                image_masks = (np.True_, np.True_, np.True_)
+            case _:
+                raise ValueError(f"Unsupported model type: {self.model_type}")
+
+        inputs = {
+            "state": state,
+            "image": dict(zip(names, images, strict=True)),
+            "image_mask": dict(zip(names, image_masks, strict=True)),
+        }
+
+        if "actions" in data:
+            inputs["actions"] = np.asarray(data["actions"])
+
+        if "prompt" in data:
+            if isinstance(data["prompt"], bytes):
+                data["prompt"] = data["prompt"].decode("utf-8")
+            inputs["prompt"] = data["prompt"]
+
+        return inputs
+
+
+@dataclasses.dataclass(frozen=True)
+class DroidOutputs(transforms.DataTransformFn):
+    def __call__(self, data: dict) -> dict:
+        # Only return the first 8 dims.
+        return {"actions": np.asarray(data["actions"][:, :8])}

real_inference/source/XPolicyLab/policy/Pi_05/openpi/src/openpi/policies/libero_policy.py

@@ -0,0 +1,100 @@
+import dataclasses
+
+import einops
+import numpy as np
+
+from openpi import transforms
+from openpi.models import model as _model
+
+
+def make_libero_example() -> dict:
+    """Creates a random input example for the Libero policy."""
+    return {
+        "observation/state": np.random.rand(8),
+        "observation/image": np.random.randint(256, size=(224, 224, 3), dtype=np.uint8),
+        "observation/wrist_image": np.random.randint(256, size=(224, 224, 3), dtype=np.uint8),
+        "prompt": "do something",
+    }
+
+
+def _parse_image(image) -> np.ndarray:
+    image = np.asarray(image)
+    if np.issubdtype(image.dtype, np.floating):
+        image = (255 * image).astype(np.uint8)
+    if image.shape[0] == 3:
+        image = einops.rearrange(image, "c h w -> h w c")
+    return image
+
+
+@dataclasses.dataclass(frozen=True)
+class LiberoInputs(transforms.DataTransformFn):
+    """
+    This class is used to convert inputs to the model to the expected format. It is used for both training and inference.
+
+    For your own dataset, you can copy this class and modify the keys based on the comments below to pipe
+    the correct elements of your dataset into the model.
+    """
+
+    # Determines which model will be used.
+    # Do not change this for your own dataset.
+    model_type: _model.ModelType
+
+    def __call__(self, data: dict) -> dict:
+        # Possibly need to parse images to uint8 (H,W,C) since LeRobot automatically
+        # stores as float32 (C,H,W), gets skipped for policy inference.
+        # Keep this for your own dataset, but if your dataset stores the images
+        # in a different key than "observation/image" or "observation/wrist_image",
+        # you should change it below.
+        # Pi0 models support three image inputs at the moment: one third-person view,
+        # and two wrist views (left and right). If your dataset does not have a particular type
+        # of image, e.g. wrist images, you can comment it out here and replace it with zeros like we do for the
+        # right wrist image below.
+        base_image = _parse_image(data["observation/image"])
+        wrist_image = _parse_image(data["observation/wrist_image"])
+
+        # Create inputs dict. Do not change the keys in the dict below.
+        inputs = {
+            "state": data["observation/state"],
+            "image": {
+                "base_0_rgb": base_image,
+                "left_wrist_0_rgb": wrist_image,
+                # Pad any non-existent images with zero-arrays of the appropriate shape.
+                "right_wrist_0_rgb": np.zeros_like(base_image),
+            },
+            "image_mask": {
+                "base_0_rgb": np.True_,
+                "left_wrist_0_rgb": np.True_,
+                # We only mask padding images for pi0 model, not pi0-FAST. Do not change this for your own dataset.
+                "right_wrist_0_rgb": np.True_ if self.model_type == _model.ModelType.PI0_FAST else np.False_,
+            },
+        }
+
+        # Pad actions to the model action dimension. Keep this for your own dataset.
+        # Actions are only available during training.
+        if "actions" in data:
+            inputs["actions"] = data["actions"]
+
+        # Pass the prompt (aka language instruction) to the model.
+        # Keep this for your own dataset (but modify the key if the instruction is not
+        # stored in "prompt"; the output dict always needs to have the key "prompt").
+        if "prompt" in data:
+            inputs["prompt"] = data["prompt"]
+
+        return inputs
+
+
+@dataclasses.dataclass(frozen=True)
+class LiberoOutputs(transforms.DataTransformFn):
+    """
+    This class is used to convert outputs from the model back the the dataset specific format. It is
+    used for inference only.
+
+    For your own dataset, you can copy this class and modify the action dimension based on the comments below.
+    """
+
+    def __call__(self, data: dict) -> dict:
+        # Only return the first N actions -- since we padded actions above to fit the model action
+        # dimension, we need to now parse out the correct number of actions in the return dict.
+        # For Libero, we only return the first 7 actions (since the rest is padding).
+        # For your own dataset, replace `7` with the action dimension of your dataset.
+        return {"actions": np.asarray(data["actions"][:, :7])}

real_inference/source/XPolicyLab/policy/Pi_05/openpi/src/openpi/policies/pika_bimanual_policy.py

@@ -0,0 +1,163 @@
+"""Policy transforms for the dual-arm Pika T-shirt-folding dataset.
+
+The LeRobot conversion stores two RGB streams only: ``cam_left_wrist`` is the
+left Pika fisheye and ``cam_right_wrist`` is the right Pika fisheye. Pi0.5 has
+three fixed image slots, so the unused third slot is explicitly masked out.
+
+The raw 14-D state/action convention is left [xyz, rotvec_xyz, gripper]
+followed by right [xyz, rotvec_xyz, gripper], with absolute next-step targets.
+``PikaBimanualInputs(relative_pose=True)`` instead maps state/action into the
+current-frame SE(3) representation used by the relative Pi0.5 configuration.
+"""
+
+import dataclasses
+
+import einops
+import numpy as np
+from scipy.spatial.transform import Rotation
+
+from openpi import transforms
+from openpi.models import model as _model
+
+
+PIKA_BIMANUAL_ACTION_DIM = 14
+
+
+def _pose_matrix(pose: np.ndarray) -> np.ndarray:
+    """Convert one Pika ``xyz + rotvec`` pose to a homogeneous matrix."""
+    pose = np.asarray(pose, dtype=np.float32)
+    if pose.shape != (6,):
+        raise ValueError(f"Expected one 6-D Pika pose, got {pose.shape}.")
+    matrix = np.eye(4, dtype=np.float32)
+    matrix[:3, :3] = Rotation.from_rotvec(pose[3:]).as_matrix().astype(np.float32)
+    matrix[:3, 3] = pose[:3]
+    return matrix
+
+
+def _pose_vector(matrix: np.ndarray) -> np.ndarray:
+    """Convert a homogeneous matrix to Pika ``xyz + rotvec`` pose."""
+    matrix = np.asarray(matrix, dtype=np.float32)
+    if matrix.shape != (4, 4):
+        raise ValueError(f"Expected a 4x4 pose matrix, got {matrix.shape}.")
+    return np.concatenate(
+        (matrix[:3, 3], Rotation.from_matrix(matrix[:3, :3]).as_rotvec().astype(np.float32))
+    ).astype(np.float32)
+
+
+def absolute_state_to_relative(state: np.ndarray) -> np.ndarray:
+    """Express dual-Pika proprioception in the current left-Pika frame.
+
+    The left pose is deliberately identity. The right pose remains available as
+    ``T_left^-1 @ T_right`` so the policy can coordinate both hands without
+    learning a Lighthouse/world origin. Grippers remain absolute openness.
+    """
+    state = np.asarray(state, dtype=np.float32)
+    if state.shape != (PIKA_BIMANUAL_ACTION_DIM,):
+        raise ValueError(f"Expected a {PIKA_BIMANUAL_ACTION_DIM}-D Pika state, got shape {state.shape}.")
+    left_from_right = np.linalg.inv(_pose_matrix(state[:6])) @ _pose_matrix(state[7:13])
+    relative = np.zeros_like(state)
+    relative[6] = state[6]
+    relative[7:13] = _pose_vector(left_from_right)
+    relative[13] = state[13]
+    return relative
+
+
+def absolute_actions_to_relative(state: np.ndarray, actions: np.ndarray) -> np.ndarray:
+    """Express absolute future Pika targets in each arm's current local frame.
+
+    For arm ``i`` and every action-horizon step ``k`` this returns
+    ``inverse(T_i,current) @ T_i,target[k]``. This is a full SE(3)
+    representation: translation is in the current local frame and rotation is
+    a local relative rotation. Gripper values intentionally remain absolute
+    normalized openings.
+    """
+    state = np.asarray(state, dtype=np.float32)
+    actions = np.asarray(actions, dtype=np.float32)
+    if state.shape != (PIKA_BIMANUAL_ACTION_DIM,):
+        raise ValueError(f"Expected a {PIKA_BIMANUAL_ACTION_DIM}-D Pika state, got shape {state.shape}.")
+    if actions.ndim != 2 or actions.shape[1] != PIKA_BIMANUAL_ACTION_DIM:
+        raise ValueError(
+            f"Expected Pika actions with shape (horizon, {PIKA_BIMANUAL_ACTION_DIM}), got {actions.shape}."
+        )
+    relative = actions.copy()
+    for offset in (0, 7):
+        current_from_world = np.linalg.inv(_pose_matrix(state[offset : offset + 6]))
+        for index, target in enumerate(actions[:, offset : offset + 6]):
+            relative[index, offset : offset + 6] = _pose_vector(current_from_world @ _pose_matrix(target))
+    return relative
+
+
+def _parse_image(image: np.ndarray) -> np.ndarray:
+    """Return an already-decoded image as uint8 HWC, accepting CHW or HWC."""
+    image = np.asarray(image)
+    if image.ndim != 3:
+        raise ValueError(f"Expected an RGB image with 3 dimensions, got shape {image.shape}.")
+    if np.issubdtype(image.dtype, np.floating):
+        image = np.clip(image * 255, 0, 255).astype(np.uint8)
+    if image.shape[0] == 3 and image.shape[-1] != 3:
+        image = einops.rearrange(image, "c h w -> h w c")
+    if image.shape[-1] != 3:
+        raise ValueError(f"Expected an RGB image with 3 channels, got shape {image.shape}.")
+    return image
+
+
+@dataclasses.dataclass(frozen=True)
+class PikaBimanualInputs(transforms.DataTransformFn):
+    """Map Pika training/runtime inputs to Pi0.5's fixed observation schema."""
+
+    model_type: _model.ModelType
+    relative_pose: bool = False
+
+    def __call__(self, data: dict) -> dict:
+        state = np.asarray(data["observation/state"], dtype=np.float32)
+        if state.shape[-1] != PIKA_BIMANUAL_ACTION_DIM:
+            raise ValueError(
+                f"Expected a {PIKA_BIMANUAL_ACTION_DIM}-D Pika state, got shape {state.shape}."
+            )
+
+        left_fisheye = _parse_image(data["observation/image"])
+        right_fisheye = _parse_image(data["observation/wrist_image"])
+        if left_fisheye.shape != right_fisheye.shape:
+            raise ValueError(
+                "The two Pika fisheye frames must have the same shape; got "
+                f"{left_fisheye.shape} and {right_fisheye.shape}."
+            )
+
+        inputs = {
+            "state": absolute_state_to_relative(state) if self.relative_pose else state,
+            "image": {
+                "base_0_rgb": left_fisheye,
+                "left_wrist_0_rgb": right_fisheye,
+                "right_wrist_0_rgb": np.zeros_like(left_fisheye),
+            },
+            "image_mask": {
+                "base_0_rgb": np.True_,
+                "left_wrist_0_rgb": np.True_,
+                "right_wrist_0_rgb": np.False_,
+            },
+        }
+
+        if "actions" in data:
+            actions = np.asarray(data["actions"], dtype=np.float32)
+            if actions.shape[-1] != PIKA_BIMANUAL_ACTION_DIM:
+                raise ValueError(
+                    f"Expected {PIKA_BIMANUAL_ACTION_DIM}-D Pika actions, got shape {actions.shape}."
+                )
+            inputs["actions"] = absolute_actions_to_relative(state, actions) if self.relative_pose else actions
+        if "prompt" in data:
+            inputs["prompt"] = data["prompt"]
+        return inputs
+
+
+@dataclasses.dataclass(frozen=True)
+class PikaBimanualOutputs(transforms.DataTransformFn):
+    """Validate and return the 14-D Pika action chunk at policy inference."""
+
+    def __call__(self, data: dict) -> dict:
+        actions = np.asarray(data["actions"])
+        if actions.shape[-1] != PIKA_BIMANUAL_ACTION_DIM:
+            raise ValueError(
+                f"Expected model output with {PIKA_BIMANUAL_ACTION_DIM} action dimensions, "
+                f"got shape {actions.shape}."
+            )
+        return {"actions": actions}

real_inference/source/XPolicyLab/policy/Pi_05/openpi/src/openpi/policies/pika_bimanual_policy_test.py

@@ -0,0 +1,44 @@
+import numpy as np
+
+from openpi.models import model as _model
+from openpi.policies import pika_bimanual_policy
+
+
+def test_pika_bimanual_inputs_maps_two_fisheyes_and_masks_third_slot():
+    transform = pika_bimanual_policy.PikaBimanualInputs(_model.ModelType.PI05)
+    result = transform(
+        {
+            "observation/state": np.zeros(14, dtype=np.float32),
+            "observation/image": np.zeros((3, 4, 5), dtype=np.uint8),
+            "observation/wrist_image": np.ones((4, 5, 3), dtype=np.uint8),
+            "actions": np.zeros((16, 14), dtype=np.float32),
+            "prompt": "Fold the T-shirt.",
+        }
+    )
+
+    assert result["image"]["base_0_rgb"].shape == (4, 5, 3)
+    assert result["image"]["left_wrist_0_rgb"].shape == (4, 5, 3)
+    assert not result["image_mask"]["right_wrist_0_rgb"]
+    assert result["actions"].shape == (16, 14)
+
+
+def test_pika_bimanual_outputs_rejects_wrong_action_dimension():
+    transform = pika_bimanual_policy.PikaBimanualOutputs()
+    with np.testing.assert_raises_regex(ValueError, "14"):
+        transform({"actions": np.zeros((16, 15), dtype=np.float32)})
+
+
+def test_relative_pika_pose_transform_removes_global_frame_and_preserves_arm_relation():
+    state = np.array([1, 2, 3, 0, 0, np.pi / 2, 0.2, 2, 2, 3, 0, 0, np.pi / 2, 0.8], dtype=np.float32)
+    actions = np.stack((state.copy(), state.copy()))
+    actions[1, 0] += 0.1  # Pika-world x, which is current local y after the z rotation.
+
+    relative_state = pika_bimanual_policy.absolute_state_to_relative(state)
+    relative_actions = pika_bimanual_policy.absolute_actions_to_relative(state, actions)
+
+    np.testing.assert_allclose(relative_state[:6], 0.0, atol=1e-6)
+    np.testing.assert_allclose(relative_state[7:10], [0.0, -1.0, 0.0], atol=1e-6)
+    np.testing.assert_allclose(relative_state[[6, 13]], [0.2, 0.8], atol=1e-6)
+    np.testing.assert_allclose(relative_actions[0, :6], 0.0, atol=1e-6)
+    np.testing.assert_allclose(relative_actions[1, :3], [0.0, -0.1, 0.0], atol=1e-6)
+    np.testing.assert_allclose(relative_actions[:, [6, 13]], [[0.2, 0.8], [0.2, 0.8]], atol=1e-6)

real_inference/source/XPolicyLab/policy/Pi_05/openpi/src/openpi/policies/policy.py

@@ -0,0 +1,156 @@
+from collections.abc import Sequence
+import logging
+import pathlib
+import time
+from typing import Any, TypeAlias
+
+import flax
+import flax.traverse_util
+import jax
+import jax.numpy as jnp
+import numpy as np
+from openpi_client import base_policy as _base_policy
+import torch
+from typing_extensions import override
+
+from openpi import transforms as _transforms
+from openpi.models import model as _model
+from openpi.shared import array_typing as at
+from openpi.shared import nnx_utils
+
+BasePolicy: TypeAlias = _base_policy.BasePolicy
+
+
+class Policy(BasePolicy):
+    def __init__(
+        self,
+        model: _model.BaseModel,
+        *,
+        rng: at.KeyArrayLike | None = None,
+        transforms: Sequence[_transforms.DataTransformFn] = (),
+        output_transforms: Sequence[_transforms.DataTransformFn] = (),
+        sample_kwargs: dict[str, Any] | None = None,
+        metadata: dict[str, Any] | None = None,
+        pytorch_device: str = "cpu",
+        is_pytorch: bool = False,
+    ):
+        """Initialize the Policy.
+
+        Args:
+            model: The model to use for action sampling.
+            rng: Random number generator key for JAX models. Ignored for PyTorch models.
+            transforms: Input data transformations to apply before inference.
+            output_transforms: Output data transformations to apply after inference.
+            sample_kwargs: Additional keyword arguments to pass to model.sample_actions.
+            metadata: Additional metadata to store with the policy.
+            pytorch_device: Device to use for PyTorch models (e.g., "cpu", "cuda:0").
+                          Only relevant when is_pytorch=True.
+            is_pytorch: Whether the model is a PyTorch model. If False, assumes JAX model.
+        """
+        self._model = model
+        self._input_transform = _transforms.compose(transforms)
+        self._output_transform = _transforms.compose(output_transforms)
+        self._sample_kwargs = sample_kwargs or {}
+        self._metadata = metadata or {}
+        self._is_pytorch_model = is_pytorch
+        self._pytorch_device = pytorch_device
+
+        if self._is_pytorch_model:
+            self._model = self._model.to(pytorch_device)
+            self._model.eval()
+            self._sample_actions = model.sample_actions
+        else:
+            # JAX model setup
+            self._sample_actions = nnx_utils.module_jit(model.sample_actions)
+            self._rng = rng or jax.random.key(0)
+
+    @override
+    def infer(self, obs: dict, *, noise: np.ndarray | None = None) -> dict:  # type: ignore[misc]
+        # Make a copy since transformations may modify the inputs in place.
+        inputs = jax.tree.map(lambda x: x, obs)
+        inputs = self._input_transform(inputs)
+
+        is_batched = False
+        if "state" in inputs:
+            is_batched = np.asarray(inputs["state"]).ndim > 1
+        elif "image" in inputs and inputs["image"]:
+            first_image = next(iter(inputs["image"].values()))
+            is_batched = np.asarray(first_image).ndim > 3
+
+        if not self._is_pytorch_model:
+            if not is_batched:
+                inputs = jax.tree.map(lambda x: jnp.asarray(x)[np.newaxis, ...], inputs)
+            else:
+                inputs = jax.tree.map(jnp.asarray, inputs)
+            self._rng, sample_rng_or_pytorch_device = jax.random.split(self._rng)
+        else:
+            # Convert inputs to PyTorch tensors and move to correct device
+            if not is_batched:
+                inputs = jax.tree.map(
+                    lambda x: torch.from_numpy(np.array(x)).to(self._pytorch_device)[None, ...], inputs
+                )
+            else:
+                inputs = jax.tree.map(lambda x: torch.from_numpy(np.array(x)).to(self._pytorch_device), inputs)
+            sample_rng_or_pytorch_device = self._pytorch_device
+
+        # Prepare kwargs for sample_actions
+        sample_kwargs = dict(self._sample_kwargs)
+        if noise is not None:
+            noise = torch.from_numpy(noise).to(self._pytorch_device) if self._is_pytorch_model else jnp.asarray(noise)
+
+            if not is_batched and noise.ndim == 2:
+                noise = noise[None, ...]
+            sample_kwargs["noise"] = noise
+
+        observation = _model.Observation.from_dict(inputs)
+        start_time = time.monotonic()
+        outputs = {
+            "state": inputs["state"],
+            "actions": self._sample_actions(sample_rng_or_pytorch_device, observation, **sample_kwargs),
+        }
+        model_time = time.monotonic() - start_time
+        if self._is_pytorch_model:
+            if not is_batched:
+                outputs = jax.tree.map(lambda x: np.asarray(x[0, ...].detach().cpu()), outputs)
+            else:
+                outputs = jax.tree.map(lambda x: np.asarray(x.detach().cpu()), outputs)
+        else:
+            if not is_batched:
+                outputs = jax.tree.map(lambda x: np.asarray(x[0, ...]), outputs)
+            else:
+                outputs = jax.tree.map(np.asarray, outputs)
+
+        outputs = self._output_transform(outputs)
+        outputs["policy_timing"] = {
+            "infer_ms": model_time * 1000,
+        }
+        return outputs
+
+    @property
+    def metadata(self) -> dict[str, Any]:
+        return self._metadata
+
+
+class PolicyRecorder(_base_policy.BasePolicy):
+    """Records the policy's behavior to disk."""
+
+    def __init__(self, policy: _base_policy.BasePolicy, record_dir: str):
+        self._policy = policy
+
+        logging.info(f"Dumping policy records to: {record_dir}")
+        self._record_dir = pathlib.Path(record_dir)
+        self._record_dir.mkdir(parents=True, exist_ok=True)
+        self._record_step = 0
+
+    @override
+    def infer(self, obs: dict) -> dict:  # type: ignore[misc]
+        results = self._policy.infer(obs)
+
+        data = {"inputs": obs, "outputs": results}
+        data = flax.traverse_util.flatten_dict(data, sep="/")
+
+        output_path = self._record_dir / f"step_{self._record_step}"
+        self._record_step += 1
+
+        np.save(output_path, np.asarray(data))
+        return results

real_inference/source/XPolicyLab/policy/Pi_05/openpi/src/openpi/policies/policy_config.py

@@ -0,0 +1,94 @@
+import logging
+import os
+import pathlib
+from typing import Any
+
+import jax.numpy as jnp
+
+import openpi.models.model as _model
+import openpi.policies.policy as _policy
+import openpi.shared.download as download
+from openpi.training import checkpoints as _checkpoints
+from openpi.training import config as _config
+import openpi.transforms as transforms
+
+
+def create_trained_policy(
+    train_config: _config.TrainConfig,
+    checkpoint_dir: pathlib.Path | str,
+    *,
+    repack_transforms: transforms.Group | None = None,
+    sample_kwargs: dict[str, Any] | None = None,
+    default_prompt: str | None = None,
+    norm_stats: dict[str, transforms.NormStats] | None = None,
+    pytorch_device: str | None = None,
+) -> _policy.Policy:
+    """Create a policy from a trained checkpoint.
+
+    Args:
+        train_config: The training config to use to create the model.
+        checkpoint_dir: The directory to load the model from.
+        repack_transforms: Optional transforms that will be applied before any other transforms.
+        sample_kwargs: The kwargs to pass to the `sample_actions` method. If not provided, the default
+            kwargs will be used.
+        default_prompt: The default prompt to use for the policy. Will inject the prompt into the input
+            data if it doesn't already exist.
+        norm_stats: The norm stats to use for the policy. If not provided, the norm stats will be loaded
+            from the checkpoint directory.
+        pytorch_device: Device to use for PyTorch models (e.g., "cpu", "cuda", "cuda:0").
+                      If None and is_pytorch=True, will use "cuda" if available, otherwise "cpu".
+
+    Note:
+        The function automatically detects whether the model is PyTorch-based by checking for the
+        presence of "model.safensors" in the checkpoint directory.
+    """
+    repack_transforms = repack_transforms or transforms.Group()
+    checkpoint_dir = download.maybe_download(str(checkpoint_dir))
+
+    # Check if this is a PyTorch model by looking for model.safetensors
+    weight_path = os.path.join(checkpoint_dir, "model.safetensors")
+    is_pytorch = os.path.exists(weight_path)
+
+    logging.info("Loading model...")
+    if is_pytorch:
+        model = train_config.model.load_pytorch(train_config, weight_path)
+        model.paligemma_with_expert.to_bfloat16_for_selected_params("bfloat16")
+    else:
+        model = train_config.model.load(_model.restore_params(checkpoint_dir / "params", dtype=jnp.bfloat16))
+    data_config = train_config.data.create(train_config.assets_dirs, train_config.model)
+    if norm_stats is None:
+        # We are loading the norm stats from the checkpoint instead of the config assets dir to make sure
+        # that the policy is using the same normalization stats as the original training process.
+        if data_config.asset_id is None:
+            raise ValueError("Asset id is required to load norm stats.")
+        norm_stats = _checkpoints.load_norm_stats(checkpoint_dir / "assets", data_config.asset_id)
+
+    # Determine the device to use for PyTorch models
+    if is_pytorch and pytorch_device is None:
+        try:
+            import torch
+
+            pytorch_device = "cuda" if torch.cuda.is_available() else "cpu"
+        except ImportError:
+            pytorch_device = "cpu"
+
+    return _policy.Policy(
+        model,
+        transforms=[
+            *repack_transforms.inputs,
+            transforms.InjectDefaultPrompt(default_prompt),
+            *data_config.data_transforms.inputs,
+            transforms.Normalize(norm_stats, use_quantiles=data_config.use_quantile_norm),
+            *data_config.model_transforms.inputs,
+        ],
+        output_transforms=[
+            *data_config.model_transforms.outputs,
+            transforms.Unnormalize(norm_stats, use_quantiles=data_config.use_quantile_norm),
+            *data_config.data_transforms.outputs,
+            *repack_transforms.outputs,
+        ],
+        sample_kwargs=sample_kwargs,
+        metadata=train_config.policy_metadata,
+        is_pytorch=is_pytorch,
+        pytorch_device=pytorch_device if is_pytorch else None,
+    )

real_inference/source/XPolicyLab/policy/Pi_05/openpi/src/openpi/policies/policy_test.py

@@ -0,0 +1,34 @@
+from openpi_client import action_chunk_broker
+import pytest
+
+from openpi.policies import aloha_policy
+from openpi.policies import policy_config as _policy_config
+from openpi.training import config as _config
+
+
+@pytest.mark.manual
+def test_infer():
+    config = _config.get_config("pi0_aloha_sim")
+    policy = _policy_config.create_trained_policy(config, "gs://openpi-assets/checkpoints/pi0_aloha_sim")
+
+    example = aloha_policy.make_aloha_example()
+    result = policy.infer(example)
+
+    assert result["actions"].shape == (config.model.action_horizon, 14)
+
+
+@pytest.mark.manual
+def test_broker():
+    config = _config.get_config("pi0_aloha_sim")
+    policy = _policy_config.create_trained_policy(config, "gs://openpi-assets/checkpoints/pi0_aloha_sim")
+
+    broker = action_chunk_broker.ActionChunkBroker(
+        policy,
+        # Only execute the first half of the chunk.
+        action_horizon=config.model.action_horizon // 2,
+    )
+
+    example = aloha_policy.make_aloha_example()
+    for _ in range(config.model.action_horizon):
+        outputs = broker.infer(example)
+        assert outputs["actions"].shape == (14,)

real_inference/source/XPolicyLab/policy/Pi_05/openpi/src/openpi/policies/wuji_policy.py

@@ -0,0 +1,88 @@
+"""Policy transforms for Tianji Marvin + dual Wuji Hand (Gen1).
+
+Adapted from wuji-openpi. Action layout (54D):
+  left arm (7) + left hand (20) + right arm (7) + right hand (20)
+
+Expected LeRobot fields:
+- observation.state: 54 dims
+- action: 54 dims
+- observation.images.cam_left_wrist / cam_right_wrist
+- observation.images.stereo_right (or cam_high via DataConfig repack)
+"""
+
+import dataclasses
+
+import einops
+import numpy as np
+
+from openpi import transforms
+from openpi.models import model as _model
+
+# Dual-arm + dual Wuji Hand Gen1
+WUJI_ACTION_DIM = 54
+
+
+def make_wuji_example() -> dict:
+    """Creates a random input example for the Wuji policy."""
+    return {
+        "observation/state": np.random.rand(WUJI_ACTION_DIM).astype(np.float32),
+        "observation/image": np.random.randint(256, size=(480, 640, 3), dtype=np.uint8),
+        "observation/left_wrist_image": np.random.randint(256, size=(480, 640, 3), dtype=np.uint8),
+        "observation/right_wrist_image": np.random.randint(256, size=(480, 640, 3), dtype=np.uint8),
+        "prompt": "pick up the object with both hands",
+    }
+
+
+def _parse_image(image) -> np.ndarray:
+    """Parse image to uint8 (H, W, C) format."""
+    image = np.asarray(image)
+    if np.issubdtype(image.dtype, np.floating):
+        image = (255 * image).astype(np.uint8)
+    if image.shape[0] == 3:
+        image = einops.rearrange(image, "c h w -> h w c")
+    return image
+
+
+@dataclasses.dataclass(frozen=True)
+class WujiInputs(transforms.DataTransformFn):
+    """Convert Wuji dataset / runtime inputs to Pi0 model format."""
+
+    model_type: _model.ModelType
+
+    def __call__(self, data: dict) -> dict:
+        base_image = _parse_image(data["observation/image"])
+        left_wrist_image = _parse_image(data["observation/left_wrist_image"])
+        right_wrist_image = _parse_image(data["observation/right_wrist_image"])
+
+        inputs = {
+            "state": data["observation/state"],
+            "image": {
+                "base_0_rgb": base_image,
+                "left_wrist_0_rgb": left_wrist_image,
+                "right_wrist_0_rgb": right_wrist_image,
+            },
+            "image_mask": {
+                "base_0_rgb": np.True_,
+                "left_wrist_0_rgb": np.True_,
+                "right_wrist_0_rgb": np.True_,
+            },
+        }
+
+        if "actions" in data:
+            inputs["actions"] = data["actions"]
+
+        if "prompt" in data:
+            inputs["prompt"] = data["prompt"]
+
+        return inputs
+
+
+@dataclasses.dataclass(frozen=True)
+class WujiOutputs(transforms.DataTransformFn):
+    """Trim padded model actions back to the Wuji action dimension."""
+
+    action_dim: int = WUJI_ACTION_DIM
+
+    def __call__(self, data: dict) -> dict:
+        actions = np.asarray(data["actions"])
+        return {"actions": actions[:, : self.action_dim]}

real_inference/source/XPolicyLab/policy/Pi_05/openpi/src/openpi/training/checkpoints.py

@@ -0,0 +1,159 @@
+from __future__ import annotations
+
+import asyncio
+import concurrent.futures as futures
+import dataclasses
+import logging
+from typing import Protocol
+
+from etils import epath
+import jax
+import orbax.checkpoint as ocp
+import orbax.checkpoint.future as future
+
+from openpi.shared import array_typing as at
+import openpi.shared.normalize as _normalize
+import openpi.training.data_loader as _data_loader
+import openpi.training.utils as training_utils
+
+
+def initialize_checkpoint_dir(
+    checkpoint_dir: epath.Path | str, *, keep_period: int | None, overwrite: bool, resume: bool
+) -> tuple[ocp.CheckpointManager, bool]:
+    checkpoint_dir = epath.Path(checkpoint_dir).resolve()
+    resuming = False
+    if checkpoint_dir.exists():
+        if overwrite:
+            checkpoint_dir.rmtree()
+            checkpoint_dir.mkdir(parents=True, exist_ok=True)
+            logging.info(f"Wiped checkpoint directory {checkpoint_dir}")
+        elif resume:
+            resuming = True
+        else:
+            raise FileExistsError(
+                f"Checkpoint directory {checkpoint_dir} already exists. Use --overwrite or --resume "
+                "to indicate how to handle it."
+            )
+
+    checkpoint_dir.mkdir(parents=True, exist_ok=True)
+
+    mngr = ocp.CheckpointManager(
+        checkpoint_dir,
+        item_handlers={
+            "assets": CallbackHandler(),
+            "train_state": ocp.PyTreeCheckpointHandler(),
+            "params": ocp.PyTreeCheckpointHandler(),
+        },
+        options=ocp.CheckpointManagerOptions(
+            max_to_keep=1,
+            keep_period=keep_period,
+            create=False,
+            async_options=ocp.AsyncOptions(timeout_secs=7200),
+        ),
+    )
+
+    # Special case: the checkpoint directory exists and the user requests to resume training, but the training run did
+    # not get to the first checkpoint saved. In this case, we don't actually want the train script to try and restore a
+    # checkpoint, since it will fail.
+    if resuming and tuple(mngr.all_steps()) in [(), (0,)]:
+        logging.info("Checkpoint directory exists, but does not contain any checkpoints. Aborting resume.")
+        resuming = False
+
+    return mngr, resuming
+
+
+def save_state(
+    checkpoint_manager: ocp.CheckpointManager,
+    state: training_utils.TrainState,
+    data_loader: _data_loader.DataLoader,
+    step: int,
+):
+    def save_assets(directory: epath.Path):
+        # Save the normalization stats.
+        data_config = data_loader.data_config()
+        norm_stats = data_config.norm_stats
+        if norm_stats is not None and data_config.asset_id is not None:
+            _normalize.save(directory / data_config.asset_id, norm_stats)
+
+    # Split params that can be used for inference into a separate item.
+    with at.disable_typechecking():
+        train_state, params = _split_params(state)
+    items = {
+        "assets": save_assets,
+        "train_state": train_state,
+        "params": {"params": params},
+    }
+    checkpoint_manager.save(step, items)
+
+
+def restore_state(
+    checkpoint_manager: ocp.CheckpointManager,
+    state: training_utils.TrainState,
+    data_loader: _data_loader.DataLoader,
+    step: int | None = None,
+) -> training_utils.TrainState:
+    del data_loader
+
+    with at.disable_typechecking():
+        # Split params that can be used for inference into a separate item.
+        train_state, params = _split_params(state)
+        restored = checkpoint_manager.restore(
+            step,
+            items={
+                "train_state": train_state,
+                "params": {"params": params},
+            },
+        )
+    return _merge_params(restored["train_state"], restored["params"])
+
+
+def load_norm_stats(assets_dir: epath.Path | str, asset_id: str) -> dict[str, _normalize.NormStats] | None:
+    norm_stats_dir = epath.Path(assets_dir) / asset_id
+    norm_stats = _normalize.load(norm_stats_dir)
+    logging.info(f"Loaded norm stats from {norm_stats_dir}")
+    return norm_stats
+
+
+class Callback(Protocol):
+    def __call__(self, directory: epath.Path) -> None: ...
+
+
+class CallbackHandler(ocp.AsyncCheckpointHandler):
+    """A CheckpointHandler for calling an arbitrary function asynchronously. Only for saving, not for restoring."""
+
+    def save(self, directory: epath.Path, args: CallbackSave):
+        if jax.process_index() == 0:
+            args.callback(directory)
+
+    async def async_save(self, directory: epath.Path, args: CallbackSave) -> list[futures.Future]:
+        return [future.CommitFutureAwaitingContractedSignals(asyncio.to_thread(self.save, directory, args))]
+
+    def restore(self, *args, **kwargs):
+        raise NotImplementedError("CallbackHandler does not support restore")
+
+
+@ocp.args.register_with_handler(CallbackHandler, for_save=True)
+@dataclasses.dataclass
+class CallbackSave(ocp.args.CheckpointArgs):
+    callback: Callback
+
+
+@ocp.args.register_with_handler(CallbackHandler, for_restore=True)
+class CallbackRestore(ocp.args.CheckpointArgs): ...
+
+
+def _split_params(state: training_utils.TrainState) -> tuple[training_utils.TrainState, at.Params]:
+    if state.ema_params is not None:
+        params = state.ema_params
+        train_state = dataclasses.replace(state, ema_params=None)
+    else:
+        params = state.params
+        train_state = dataclasses.replace(state, params={})
+    return train_state, params
+
+
+def _merge_params(train_state: training_utils.TrainState, params: dict[str, at.Params]) -> training_utils.TrainState:
+    # Revert the logic inside `_split_params`. Assumes that existence of `params` means that EMA params were used during the split.
+    if train_state.params:
+        return dataclasses.replace(train_state, ema_params=params["params"])
+    return dataclasses.replace(train_state, params=params["params"])

real_inference/source/XPolicyLab/policy/Pi_05/openpi/src/openpi/training/config.py

@@ -0,0 +1,1026 @@
+"""See _CONFIGS for the list of available configs."""
+
+import abc
+from collections.abc import Sequence
+import dataclasses
+import difflib
+import logging
+import os
+import pathlib
+from typing import Any, Literal, Protocol, TypeAlias
+
+import etils.epath as epath
+import flax.nnx as nnx
+from typing_extensions import override
+import tyro
+
+import openpi.models.model as _model
+import openpi.models.pi0_config as pi0_config
+import openpi.models.pi0_fast as pi0_fast
+import openpi.models.tokenizer as _tokenizer
+import openpi.policies.aloha_policy as aloha_policy
+import openpi.policies.droid_policy as droid_policy
+import openpi.policies.libero_policy as libero_policy
+import openpi.policies.pika_bimanual_policy as pika_bimanual_policy
+import openpi.policies.wuji_policy as wuji_policy
+import openpi.shared.download as _download
+import openpi.shared.nnx_utils as nnx_utils
+import openpi.shared.normalize as _normalize
+import openpi.training.droid_rlds_dataset as droid_rlds_dataset
+import openpi.training.misc.polaris_config as polaris_config
+import openpi.training.misc.roboarena_config as roboarena_config
+import openpi.training.optimizer as _optimizer
+import openpi.training.weight_loaders as weight_loaders
+import openpi.transforms as _transforms
+
+# RoboDojo normalization assets bundled with this adapter (openpi/assets/RoboDojo_assets).
+_ROBODOJO_ASSETS_DIR = pathlib.Path(__file__).resolve().parents[3] / "assets" / "RoboDojo_assets"
+# Norm stats for Tianji Marvin + Wuji Hand (written by compute_norm_stats).
+_WUJI_ASSETS_DIR = pathlib.Path(__file__).resolve().parents[3] / "assets" / "Wuji_assets"
+
+ModelType: TypeAlias = _model.ModelType
+# Work around a tyro issue with using nnx.filterlib.Filter directly.
+Filter: TypeAlias = nnx.filterlib.Filter
+
+
+@dataclasses.dataclass(frozen=True)
+class AssetsConfig:
+    """Determines the location of assets (e.g., norm stats) that will be used to set up the data pipeline.
+
+    These assets will be replicated inside the checkpoint under the `assets/asset_id` directory.
+
+    This can be used to load assets from a different checkpoint (e.g., base model checkpoint) or some other
+    centralized location. For example, to load the norm stats for the Trossen robot from the base model checkpoint
+    during fine-tuning, use:
+
+    ```
+    AssetsConfig(
+        assets_dir="gs://openpi-assets/checkpoints/pi0_base/assets",
+        asset_id="trossen",
+    )
+    ```
+    """
+
+    # Assets directory. If not provided, the config assets_dirs will be used. This is useful to load assets from
+    # a different checkpoint (e.g., base model checkpoint) or some other centralized location.
+    assets_dir: str | None = None
+
+    # Asset id. If not provided, the repo id will be used. This allows users to reference assets that describe
+    # different robot platforms.
+    asset_id: str | None = None
+
+
+@dataclasses.dataclass(frozen=True)
+class DataConfig:
+    # LeRobot repo id. If None, fake data will be created.
+    repo_id: str | None = None
+    # Optional physical root for a local LeRobot dataset. This avoids requiring
+    # the data to be placed below HF_LEROBOT_HOME/<repo_id>.
+    lerobot_root: str | None = None
+    # Optional subset of LeRobot episode indices, applied before sampling.
+    episode_indices: Sequence[int] | None = None
+    # Video decoder backend for LeRobot datasets. Forced to pyav by default because
+    # torchcodec is present in some environments but not fully functional at runtime.
+    video_backend: Literal["pyav", "torchcodec", "video_reader"] = "pyav"
+    # Directory within the assets directory containing the data assets.
+    asset_id: str | None = None
+    # Contains precomputed normalization stats. If None, normalization will not be performed.
+    norm_stats: dict[str, _transforms.NormStats] | None = None
+
+    # Used to adopt the inputs from a dataset specific format to a common format
+    # which is expected by the data transforms.
+    repack_transforms: _transforms.Group = dataclasses.field(default_factory=_transforms.Group)
+    # Data transforms, typically include robot specific transformations. Will be applied
+    # before the data is normalized. See `model.Observation` and `model.Actions` to learn about the
+    # normalized data.
+    data_transforms: _transforms.Group = dataclasses.field(default_factory=_transforms.Group)
+    # Model specific transforms. Will be applied after the data is normalized.
+    model_transforms: _transforms.Group = dataclasses.field(default_factory=_transforms.Group)
+    # If true, will use quantile normalization. Otherwise, normal z-score normalization will be used.
+    use_quantile_norm: bool = False
+
+    # Names of keys that will be used by the data loader to generate the action sequence. The length of the
+    # sequence is defined by the `action_horizon` field in the model config. This should be adjusted if your
+    # LeRobot dataset is using different keys to represent the action.
+    action_sequence_keys: Sequence[str] = ("actions",)
+
+    # If true, will use the LeRobot dataset task to define the prompt.
+    prompt_from_task: bool = False
+
+    # Only used for RLDS data loader (ie currently only used for DROID).
+    rlds_data_dir: str | None = None
+    # Action space for DROID dataset.
+    action_space: droid_rlds_dataset.DroidActionSpace | None = None
+    # List of datasets to sample from: name, version, weight, and optionally filter_dict_path
+    datasets: Sequence[droid_rlds_dataset.RLDSDataset] = ()
+
+
+class GroupFactory(Protocol):
+    def __call__(self, model_config: _model.BaseModelConfig) -> _transforms.Group:
+        """Create a group."""
+
+
+@dataclasses.dataclass(frozen=True)
+class ModelTransformFactory(GroupFactory):
+    """Creates model transforms for standard pi0 models."""
+
+    # If provided, will determine the default prompt that be used by the model.
+    default_prompt: str | None = None
+
+    def __call__(self, model_config: _model.BaseModelConfig) -> _transforms.Group:
+        match model_config.model_type:
+            case _model.ModelType.PI0:
+                return _transforms.Group(
+                    inputs=[
+                        _transforms.InjectDefaultPrompt(self.default_prompt),
+                        _transforms.ResizeImages(224, 224),
+                        _transforms.TokenizePrompt(
+                            _tokenizer.PaligemmaTokenizer(model_config.max_token_len),
+                        ),
+                        _transforms.PadStatesAndActions(model_config.action_dim),
+                    ],
+                )
+            case _model.ModelType.PI05:
+                assert isinstance(model_config, pi0_config.Pi0Config)
+                return _transforms.Group(
+                    inputs=[
+                        _transforms.InjectDefaultPrompt(self.default_prompt),
+                        _transforms.ResizeImages(224, 224),
+                        _transforms.TokenizePrompt(
+                            _tokenizer.PaligemmaTokenizer(model_config.max_token_len),
+                            discrete_state_input=model_config.discrete_state_input,
+                        ),
+                        _transforms.PadStatesAndActions(model_config.action_dim),
+                    ],
+                )
+            case _model.ModelType.PI0_FAST:
+                tokenizer_cls = (
+                    _tokenizer.FASTTokenizer
+                    if model_config.fast_model_tokenizer is None
+                    else model_config.fast_model_tokenizer
+                )
+                tokenizer_kwargs = (
+                    {} if model_config.fast_model_tokenizer_kwargs is None else model_config.fast_model_tokenizer_kwargs
+                )
+                return _transforms.Group(
+                    inputs=[
+                        _transforms.InjectDefaultPrompt(self.default_prompt),
+                        _transforms.ResizeImages(224, 224),
+                        _transforms.TokenizeFASTInputs(
+                            tokenizer_cls(model_config.max_token_len, **tokenizer_kwargs),
+                        ),
+                    ],
+                    outputs=[
+                        _transforms.ExtractFASTActions(
+                            tokenizer_cls(model_config.max_token_len, **tokenizer_kwargs),
+                            action_horizon=model_config.action_horizon,
+                            action_dim=model_config.action_dim,
+                        )
+                    ],
+                )
+
+
+@dataclasses.dataclass(frozen=True)
+class DataConfigFactory(abc.ABC):
+    # The LeRobot repo id.
+    repo_id: str = tyro.MISSING
+    # Determines how the assets will be loaded.
+    assets: AssetsConfig = dataclasses.field(default_factory=AssetsConfig)
+    # Base config that will be updated by the factory.
+    base_config: tyro.conf.Suppress[DataConfig | None] = None
+
+    @abc.abstractmethod
+    def create(self, assets_dirs: pathlib.Path, model_config: _model.BaseModelConfig) -> DataConfig:
+        """Create a data config."""
+
+    def create_base_config(self, assets_dirs: pathlib.Path, model_config: _model.BaseModelConfig) -> DataConfig:
+        repo_id = self.repo_id if self.repo_id is not tyro.MISSING else None
+        asset_id = self.assets.asset_id or repo_id
+        return dataclasses.replace(
+            self.base_config or DataConfig(),
+            repo_id=repo_id,
+            asset_id=asset_id,
+            norm_stats=self._load_norm_stats(epath.Path(self.assets.assets_dir or assets_dirs), asset_id),
+            use_quantile_norm=model_config.model_type != ModelType.PI0,
+        )
+
+    def _load_norm_stats(self, assets_dir: epath.Path, asset_id: str | None) -> dict[str, _transforms.NormStats] | None:
+        if asset_id is None:
+            return None
+        try:
+            data_assets_dir = str(assets_dir / asset_id)
+            norm_stats = _normalize.load(_download.maybe_download(data_assets_dir))
+            logging.info(f"Loaded norm stats from {data_assets_dir}")
+            return norm_stats
+        except FileNotFoundError:
+            logging.info(f"Norm stats not found in {data_assets_dir}, skipping.")
+        return None
+
+
+@dataclasses.dataclass(frozen=True)
+class FakeDataConfig(DataConfigFactory):
+    repo_id: str = "fake"
+
+    @override
+    def create(self, assets_dirs: pathlib.Path, model_config: _model.BaseModelConfig) -> DataConfig:
+        return DataConfig(repo_id=self.repo_id)
+
+
+@dataclasses.dataclass(frozen=True)
+class SimpleDataConfig(DataConfigFactory):
+    # Factory for the data transforms.
+    data_transforms: tyro.conf.Suppress[GroupFactory] = dataclasses.field(default_factory=GroupFactory)
+    # Factory for the model transforms.
+    model_transforms: tyro.conf.Suppress[GroupFactory] = dataclasses.field(default_factory=ModelTransformFactory)
+
+    @override
+    def create(self, assets_dirs: pathlib.Path, model_config: _model.BaseModelConfig) -> DataConfig:
+        return dataclasses.replace(
+            self.create_base_config(assets_dirs, model_config),
+            data_transforms=self.data_transforms(model_config),
+            model_transforms=self.model_transforms(model_config),
+        )
+
+
+@dataclasses.dataclass(frozen=True)
+class LeRobotAlohaDataConfig(DataConfigFactory):
+    # If true, will convert joint dimensions to deltas with respect to the current state before passing to the model.
+    # Gripper dimensions will remain in absolute values.
+    use_delta_joint_actions: bool = True
+    # If provided, will be injected into the input data if the "prompt" key is not present.
+    default_prompt: str | None = None
+    # If true, this will convert the joint and gripper values from the standard Aloha space to
+    # the space used by the pi internal runtime which was used to train the base model. People who
+    # use standard Aloha data should set this to true.
+    adapt_to_pi: bool = False
+
+    # Repack transforms.
+    repack_transforms: tyro.conf.Suppress[_transforms.Group] = dataclasses.field(
+        default=_transforms.Group(
+            inputs=[
+                _transforms.RepackTransform(
+                    {
+                        "images": {"cam_high": "observation.images.top"},
+                        "state": "observation.state",
+                        "actions": "action",
+                    }
+                )
+            ]
+        )
+    )
+    # Action keys that will be used to read the action sequence from the dataset.
+    action_sequence_keys: Sequence[str] = ("action",)
+
+    @override
+    def create(self, assets_dirs: pathlib.Path, model_config: _model.BaseModelConfig) -> DataConfig:
+        data_transforms = _transforms.Group(
+            inputs=[aloha_policy.AlohaInputs(adapt_to_pi=self.adapt_to_pi)],
+            outputs=[aloha_policy.AlohaOutputs(adapt_to_pi=self.adapt_to_pi)],
+        )
+        if self.use_delta_joint_actions:
+            delta_action_mask = _transforms.make_bool_mask(6, -1, 6, -1)
+            data_transforms = data_transforms.push(
+                inputs=[_transforms.DeltaActions(delta_action_mask)],
+                outputs=[_transforms.AbsoluteActions(delta_action_mask)],
+            )
+
+        model_transforms = ModelTransformFactory(default_prompt=self.default_prompt)(model_config)
+
+        return dataclasses.replace(
+            self.create_base_config(assets_dirs, model_config),
+            repack_transforms=self.repack_transforms,
+            data_transforms=data_transforms,
+            model_transforms=model_transforms,
+            action_sequence_keys=self.action_sequence_keys,
+        )
+
+
+@dataclasses.dataclass(frozen=True)
+class LeRobotLiberoDataConfig(DataConfigFactory):
+    """
+    This config is used to configure transforms that are applied at various parts of the data pipeline.
+    For your own dataset, you can copy this class and modify the transforms to match your dataset based on the
+    comments below.
+    """
+
+    extra_delta_transform: bool = False
+
+    @override
+    def create(self, assets_dirs: pathlib.Path, model_config: _model.BaseModelConfig) -> DataConfig:
+        # The repack transform is *only* applied to the data coming from the dataset,
+        # and *not* during inference. We can use it to make inputs from the dataset look
+        # as close as possible to those coming from the inference environment (e.g. match the keys).
+        # Below, we match the keys in the dataset (which we defined in the data conversion script) to
+        # the keys we use in our inference pipeline (defined in the inference script for libero).
+        # For your own dataset, first figure out what keys your environment passes to the policy server
+        # and then modify the mappings below so your dataset's keys get matched to those target keys.
+        # The repack transform simply remaps key names here.
+        repack_transform = _transforms.Group(
+            inputs=[
+                _transforms.RepackTransform(
+                    {
+                        "observation/image": "image",
+                        "observation/wrist_image": "wrist_image",
+                        "observation/state": "state",
+                        "actions": "actions",
+                        "prompt": "prompt",
+                    }
+                )
+            ]
+        )
+
+        # The data transforms are applied to the data coming from the dataset *and* during inference.
+        # Below, we define the transforms for data going into the model (``inputs``) and the transforms
+        # for data coming out of the model (``outputs``) (the latter is only used during inference).
+        # We defined these transforms in `libero_policy.py`. You can check the detailed comments there for
+        # how to modify the transforms to match your dataset. Once you created your own transforms, you can
+        # replace the transforms below with your own.
+        data_transforms = _transforms.Group(
+            inputs=[libero_policy.LiberoInputs(model_type=model_config.model_type)],
+            outputs=[libero_policy.LiberoOutputs()],
+        )
+
+        # One additional data transform: pi0 models are trained on delta actions (relative to the first
+        # state in each action chunk). IF your data has ``absolute`` actions (e.g. target joint angles)
+        # you can uncomment the following line to convert the actions to delta actions. The only exception
+        # is for the gripper actions which are always absolute.
+        # In the example below, we would apply the delta conversion to the first 6 actions (joints) and
+        # leave the 7th action (gripper) unchanged, i.e. absolute.
+        # In Libero, the raw actions in the dataset are already delta actions, so we *do not* need to
+        # apply a separate delta conversion (that's why it's commented out). Choose whether to apply this
+        # transform based on whether your dataset uses ``absolute`` or ``delta`` actions out of the box.
+
+        # LIBERO already represents actions as deltas, but we have some old Pi0 checkpoints that are trained with this
+        # extra delta transform.
+        if self.extra_delta_transform:
+            delta_action_mask = _transforms.make_bool_mask(6, -1)
+            data_transforms = data_transforms.push(
+                inputs=[_transforms.DeltaActions(delta_action_mask)],
+                outputs=[_transforms.AbsoluteActions(delta_action_mask)],
+            )
+
+        # Model transforms include things like tokenizing the prompt and action targets
+        # You do not need to change anything here for your own dataset.
+        model_transforms = ModelTransformFactory()(model_config)
+
+        # We return all data transforms for training and inference. No need to change anything here.
+        return dataclasses.replace(
+            self.create_base_config(assets_dirs, model_config),
+            repack_transforms=repack_transform,
+            data_transforms=data_transforms,
+            model_transforms=model_transforms,
+        )
+
+
+@dataclasses.dataclass(frozen=True)
+class LeRobotWujiDataConfig(DataConfigFactory):
+    """
+    Config for Tianji Marvin + dual Wuji Hand Gen1 (adapted from wuji-openpi).
+
+    Dataset features:
+    - observation.state: 54 dims (7 left arm + 20 left hand + 7 right arm + 20 right hand)
+    - action: 54 dims
+    - observation.images.cam_left_wrist / cam_right_wrist
+    - base camera: observation.images.stereo_right (default) or cam_high via base_image_key
+
+    Requires model.action_dim=54. Load pi05_base with PartialCheckpointWeightLoader.
+    """
+
+    extra_delta_transform: bool = False
+    # LeRobot key for the head / stereo base camera.
+    base_image_key: str = "observation.images.stereo_right"
+    action_sequence_keys: Sequence[str] = ("action",)
+
+    @override
+    def create(self, assets_dirs: pathlib.Path, model_config: _model.BaseModelConfig) -> DataConfig:
+        repack_transform = _transforms.Group(
+            inputs=[
+                _transforms.RepackTransform(
+                    {
+                        "observation/image": self.base_image_key,
+                        "observation/left_wrist_image": "observation.images.cam_left_wrist",
+                        "observation/right_wrist_image": "observation.images.cam_right_wrist",
+                        "observation/state": "observation.state",
+                        "actions": "action",
+                        "prompt": "prompt",
+                    }
+                )
+            ]
+        )
+
+        data_transforms = _transforms.Group(
+            inputs=[wuji_policy.WujiInputs(model_type=model_config.model_type)],
+            outputs=[wuji_policy.WujiOutputs(action_dim=model_config.action_dim)],
+        )
+
+        # Arms: delta; hands: absolute (same as wuji-openpi).
+        if self.extra_delta_transform:
+            delta_action_mask = _transforms.make_bool_mask(7, -20, 7, -20)
+            data_transforms = data_transforms.push(
+                inputs=[_transforms.DeltaActions(delta_action_mask)],
+                outputs=[_transforms.AbsoluteActions(delta_action_mask)],
+            )
+
+        model_transforms = ModelTransformFactory()(model_config)
+
+        return dataclasses.replace(
+            self.create_base_config(assets_dirs, model_config),
+            repack_transforms=repack_transform,
+            data_transforms=data_transforms,
+            model_transforms=model_transforms,
+            action_sequence_keys=self.action_sequence_keys,
+        )
+
+
+@dataclasses.dataclass(frozen=True)
+class LeRobotPikaBimanualDataConfig(DataConfigFactory):
+    """Local LeRobot configuration for two-fisheye, 14-D dual-arm Pika data.
+
+    The converter writes absolute next-state targets, therefore action deltas
+    deliberately remain disabled. ``PIKA_LEROBOT_ROOT`` points at the physical
+    local dataset root and can be overridden with ``local_dataset_root``.
+    """
+
+    local_dataset_root: str | None = None
+    episode_indices: Sequence[int] = tuple(range(50))
+    default_prompt: str = "Fold the T-shirt."
+    action_sequence_keys: Sequence[str] = ("action",)
+    # Relative mode removes the arbitrary Pika/Lighthouse world origin. State
+    # is anchored at the current left Pika and each action is a full local
+    # SE(3) transform from that arm's current pose to its future target.
+    relative_pose: bool = False
+
+    @override
+    def create(self, assets_dirs: pathlib.Path, model_config: _model.BaseModelConfig) -> DataConfig:
+        if model_config.action_dim != pika_bimanual_policy.PIKA_BIMANUAL_ACTION_DIM:
+            raise ValueError(
+                "Pika bimanual config requires model.action_dim="
+                f"{pika_bimanual_policy.PIKA_BIMANUAL_ACTION_DIM}, got {model_config.action_dim}."
+            )
+        dataset_root = self.local_dataset_root or os.environ.get("PIKA_LEROBOT_ROOT")
+        if not dataset_root:
+            raise ValueError(
+                "Set PIKA_LEROBOT_ROOT to the local converted LeRobot dataset, or set "
+                "data.local_dataset_root explicitly."
+            )
+
+        repack_transform = _transforms.Group(
+            inputs=[
+                _transforms.RepackTransform(
+                    {
+                        "observation/image": "observation.images.cam_left_wrist",
+                        "observation/wrist_image": "observation.images.cam_right_wrist",
+                        "observation/state": "observation.state",
+                        "actions": "action",
+                        "prompt": "prompt",
+                    }
+                )
+            ]
+        )
+        data_transforms = _transforms.Group(
+            inputs=[
+                pika_bimanual_policy.PikaBimanualInputs(
+                    model_type=model_config.model_type,
+                    relative_pose=self.relative_pose,
+                )
+            ],
+            outputs=[pika_bimanual_policy.PikaBimanualOutputs()],
+        )
+
+        return dataclasses.replace(
+            self.create_base_config(assets_dirs, model_config),
+            lerobot_root=dataset_root,
+            episode_indices=self.episode_indices,
+            repack_transforms=repack_transform,
+            data_transforms=data_transforms,
+            model_transforms=ModelTransformFactory(default_prompt=self.default_prompt)(model_config),
+            action_sequence_keys=self.action_sequence_keys,
+            prompt_from_task=True,
+        )
+
+
+@dataclasses.dataclass(frozen=True)
+class RLDSDroidDataConfig(DataConfigFactory):
+    """
+    Config for training on DROID, using RLDS data format (for efficient training on larger datasets).
+    """
+
+    rlds_data_dir: str | None = None
+    action_space: droid_rlds_dataset.DroidActionSpace | None = None
+
+    # Filtering options. Can pass a path to a dictionary that maps episodes to timestep ranges
+    # to tuples denoting ranges of time steps to keep (start, end). Episodes are uniquely identified with
+    # f"{recording_folderpath}--{file_path}", both of which are present in the RLDS episode metadata.
+
+    # List of datasets to sample from: name, version, weight, and optionally filter_dict_path
+    datasets: Sequence[droid_rlds_dataset.RLDSDataset] = (
+        droid_rlds_dataset.RLDSDataset(
+            name="droid",
+            version="1.0.1",
+            weight=1.0,
+            filter_dict_path="gs://openpi-assets/droid/droid_sample_ranges_v1_0_1.json",
+        ),
+    )
+
+    @override
+    def create(self, assets_dirs: pathlib.Path, model_config: _model.BaseModelConfig) -> DataConfig:
+        repack_transform = _transforms.Group(
+            inputs=[
+                _transforms.RepackTransform(
+                    {
+                        "observation/exterior_image_1_left": "observation/image",
+                        "observation/wrist_image_left": "observation/wrist_image",
+                        "observation/joint_position": "observation/joint_position",
+                        "observation/gripper_position": "observation/gripper_position",
+                        "actions": "actions",
+                        "prompt": "prompt",
+                    }
+                )
+            ]
+        )
+
+        data_transforms = _transforms.Group(
+            inputs=[droid_policy.DroidInputs(model_type=model_config.model_type)],
+            outputs=[droid_policy.DroidOutputs()],
+        )
+
+        if self.action_space == droid_rlds_dataset.DroidActionSpace.JOINT_POSITION:
+            # Data loader returns absolute joint position actions -- convert to delta actions for training.
+            delta_action_mask = _transforms.make_bool_mask(7, -1)
+            data_transforms = data_transforms.push(
+                inputs=[_transforms.DeltaActions(delta_action_mask)],
+                outputs=[_transforms.AbsoluteActions(delta_action_mask)],
+            )
+
+        model_transforms = ModelTransformFactory()(model_config)
+
+        assert self.rlds_data_dir is not None, "Need to set rlds data dir for RLDS data loader."
+
+        return dataclasses.replace(
+            self.create_base_config(assets_dirs, model_config),
+            repack_transforms=repack_transform,
+            data_transforms=data_transforms,
+            model_transforms=model_transforms,
+            rlds_data_dir=self.rlds_data_dir,
+            action_space=self.action_space,
+            datasets=self.datasets,
+        )
+
+
+@dataclasses.dataclass(frozen=True)
+class LeRobotDROIDDataConfig(DataConfigFactory):
+    """
+    Example data config for custom DROID dataset in LeRobot format.
+    To convert your custom DROID dataset (<10s of hours) to LeRobot format, see examples/droid/convert_droid_data_to_lerobot.py
+    """
+
+    @override
+    def create(self, assets_dirs: pathlib.Path, model_config: _model.BaseModelConfig) -> DataConfig:
+        repack_transform = _transforms.Group(
+            inputs=[
+                _transforms.RepackTransform(
+                    {
+                        "observation/exterior_image_1_left": "exterior_image_1_left",
+                        "observation/exterior_image_2_left": "exterior_image_2_left",
+                        "observation/wrist_image_left": "wrist_image_left",
+                        "observation/joint_position": "joint_position",
+                        "observation/gripper_position": "gripper_position",
+                        "actions": "actions",
+                        "prompt": "prompt",
+                    }
+                )
+            ]
+        )
+        # We assume joint *velocity* actions, so we should *not* apply an additional delta transform.
+        data_transforms = _transforms.Group(
+            inputs=[droid_policy.DroidInputs(model_type=model_config.model_type)],
+            outputs=[droid_policy.DroidOutputs()],
+        )
+        model_transforms = ModelTransformFactory()(model_config)
+
+        return dataclasses.replace(
+            self.create_base_config(assets_dirs, model_config),
+            repack_transforms=repack_transform,
+            data_transforms=data_transforms,
+            model_transforms=model_transforms,
+        )
+
+
+@dataclasses.dataclass(frozen=True)
+class TrainConfig:
+    # Name of the config. Must be unique. Will be used to reference this config.
+    name: tyro.conf.Suppress[str]
+    # Project name.
+    project_name: str = "openpi"
+    # Experiment name. Will be used to name the metadata and checkpoint directories.
+    exp_name: str = tyro.MISSING
+
+    # Defines the model config. Some attributes (action_dim, action_horizon, and max_token_len) are shared by all models
+    # -- see BaseModelConfig. Specific model implementations (e.g., Pi0Config) inherit from BaseModelConfig and may
+    # define additional attributes.
+    model: _model.BaseModelConfig = dataclasses.field(default_factory=pi0_config.Pi0Config)
+
+    # A weight loader can optionally load (possibly partial) weights from disk after the model is initialized.
+    weight_loader: weight_loaders.WeightLoader = dataclasses.field(default_factory=weight_loaders.NoOpWeightLoader)
+
+    # Optional path to a PyTorch checkpoint to load weights from.
+    pytorch_weight_path: str | None = None
+
+    # Precision for PyTorch training.
+    pytorch_training_precision: Literal["bfloat16", "float32"] = "bfloat16"
+
+    lr_schedule: _optimizer.LRScheduleConfig = dataclasses.field(default_factory=_optimizer.CosineDecaySchedule)
+    optimizer: _optimizer.OptimizerConfig = dataclasses.field(default_factory=_optimizer.AdamW)
+    ema_decay: float | None = 0.99
+
+    # Specifies which weights should be frozen.
+    freeze_filter: tyro.conf.Suppress[Filter] = dataclasses.field(default_factory=nnx.Nothing)
+
+    # Determines the data to be trained on.
+    data: DataConfigFactory = dataclasses.field(default_factory=FakeDataConfig)
+
+    # Base directory for config assets (e.g., norm stats).
+    assets_base_dir: str = "./assets"
+    # Base directory for checkpoints.
+    checkpoint_base_dir: str = "./checkpoints"
+    # Optional exact checkpoint directory. XPolicyLab train.sh uses this to
+    # keep policy checkpoints under policy/<name>/checkpoints/<6-tuple>.
+    checkpoint_dir_override: str | None = None
+
+    # Random seed that will be used by random generators during training.
+    seed: int = 42
+    # Global batch size.
+    batch_size: int = 32
+    # Number of workers to use for the data loader. Increasing this number will speed up data loading but
+    # will increase memory and CPU usage.
+    num_workers: int = 8
+    # Number of train steps (batches) to run.
+    num_train_steps: int = 30_000
+
+    # How often (in steps) to log training metrics.
+    log_interval: int = 100
+    # How often (in steps) to save checkpoints.
+    save_interval: int = 1000
+    # If set, any existing checkpoints matching step % keep_period == 0 will not be deleted.
+    keep_period: int | None = 5000
+
+    # If true, will overwrite the checkpoint directory if it already exists.
+    overwrite: bool = False
+    # If true, will resume training from the last checkpoint.
+    resume: bool = False
+
+    # If true, will enable wandb logging.
+    wandb_enabled: bool = True
+
+    # Used to pass metadata to the policy server.
+    policy_metadata: dict[str, Any] | None = None
+
+    # If the value is greater than 1, FSDP will be enabled and shard across number of specified devices; overall
+    # device memory will be reduced but training could potentially be slower.
+    # eg. if total device is 4 and fsdp devices is 2; then the model will shard to 2 devices and run
+    # data parallel between 2 groups of devices.
+    fsdp_devices: int = 1
+
+    @property
+    def assets_dirs(self) -> pathlib.Path:
+        """Get the assets directory for this config."""
+        return (pathlib.Path(self.assets_base_dir) / self.name).resolve()
+
+    @property
+    def checkpoint_dir(self) -> pathlib.Path:
+        """Get the checkpoint directory for this config."""
+        if not self.exp_name:
+            raise ValueError("--exp_name must be set")
+        if self.checkpoint_dir_override:
+            return pathlib.Path(self.checkpoint_dir_override).resolve()
+        return (pathlib.Path(self.checkpoint_base_dir) / self.name / self.exp_name).resolve()
+
+    @property
+    def trainable_filter(self) -> nnx.filterlib.Filter:
+        """Get the filter for the trainable parameters."""
+        return nnx.All(nnx.Param, nnx.Not(self.freeze_filter))
+
+    def __post_init__(self) -> None:
+        if self.resume and self.overwrite:
+            raise ValueError("Cannot resume and overwrite at the same time.")
+
+
+# Use `get_config` if you need to get a config by name in your code.
+_PIKA_BIMANUAL_LORA_MODEL = pi0_config.Pi0Config(
+    pi05=True,
+    action_dim=14,
+    action_horizon=16,
+    max_token_len=256,
+    # Train LoRA adapters in both the PaliGemma backbone and action expert.
+    paligemma_variant="gemma_2b_lora",
+    action_expert_variant="gemma_300m_lora",
+)
+
+_ALOHA_GARMENT_MODEL = pi0_config.Pi0Config(
+    pi05=True, paligemma_variant="gemma_2b_lora", action_expert_variant="gemma_300m_lora")
+_ALOHA_GARMENT_ROOT = "/data2/wushuang/my_projects/Aloha2-CollectData-DeformableObject/policy_training"
+
+def _aloha_garment_data(split: str) -> LeRobotAlohaDataConfig:
+    return LeRobotAlohaDataConfig(
+        repo_id=f"local/aloha_garment_{split}",
+        assets=AssetsConfig(asset_id="aloha_garment"),
+        base_config=DataConfig(lerobot_root=f"{_ALOHA_GARMENT_ROOT}/lerobot_{split}", prompt_from_task=True),
+        use_delta_joint_actions=False, adapt_to_pi=False,
+        repack_transforms=_transforms.Group(inputs=[_transforms.RepackTransform({
+            "images": {"cam_high": "observation.images.cam_high", "cam_left_wrist": "observation.images.cam_left_wrist", "cam_right_wrist": "observation.images.cam_right_wrist"},
+            "state": "observation.state", "actions": "action", "prompt": "prompt"})]),
+    )
+
+_CONFIGS = [
+    TrainConfig(
+        name="pi05_aloha_garment_lora", model=_ALOHA_GARMENT_MODEL,
+        freeze_filter=nnx.Any(
+            _ALOHA_GARMENT_MODEL.get_freeze_filter(),
+            nnx_utils.PathRegex("PaliGemma/img/.*"),
+        ),
+        ema_decay=None,
+        data=_aloha_garment_data("train"),
+        weight_loader=weight_loaders.CheckpointWeightLoader("/data2/wushuang/.cache/openpi/openpi-assets/checkpoints/pi05_base/params"),
+        batch_size=2, num_workers=2, wandb_enabled=False,
+    ),
+    TrainConfig(
+        name="pi05_base_aloha_full_sim_arx-x5_seed_0",
+        model=pi0_config.Pi0Config(pi05=True),
+        data=LeRobotAlohaDataConfig(
+            repo_id="RoboDojo_sim_arx-x5_v30",
+            assets=AssetsConfig(
+                assets_dir=str(_ROBODOJO_ASSETS_DIR),
+                asset_id="arx_x5_sim",
+            ),
+            repack_transforms=_transforms.Group(
+                inputs=[
+                    _transforms.RepackTransform(
+                        {
+                            "images": {
+                                "cam_high": "observation.images.cam_high",
+                                "cam_left_wrist": "observation.images.cam_left_wrist",
+                                "cam_right_wrist": "observation.images.cam_right_wrist",
+                            },
+                            "state": "observation.state",
+                            "actions": "action",
+                            "prompt": "prompt",
+                        }
+                    )
+                ]
+            ),
+            base_config=DataConfig(
+                prompt_from_task=True,  # Set to True for prompt by task_name
+            ),
+        ),
+        weight_loader=weight_loaders.CheckpointWeightLoader("gs://openpi-assets/checkpoints/pi05_base/params"),
+        seed=0,
+        batch_size=256,
+        fsdp_devices=2,
+        num_train_steps=60000,
+    ),
+    TrainConfig(
+        name="pi05_base_aloha_full_sim_arx-x5_seed_1",
+        model=pi0_config.Pi0Config(pi05=True),
+        data=LeRobotAlohaDataConfig(
+            repo_id="RoboDojo_sim_arx-x5_v30",
+            assets=AssetsConfig(
+                assets_dir=str(_ROBODOJO_ASSETS_DIR),
+                asset_id="arx_x5_sim",
+            ),
+            repack_transforms=_transforms.Group(
+                inputs=[
+                    _transforms.RepackTransform(
+                        {
+                            "images": {
+                                "cam_high": "observation.images.cam_high",
+                                "cam_left_wrist": "observation.images.cam_left_wrist",
+                                "cam_right_wrist": "observation.images.cam_right_wrist",
+                            },
+                            "state": "observation.state",
+                            "actions": "action",
+                            "prompt": "prompt",
+                        }
+                    )
+                ]
+            ),
+            base_config=DataConfig(
+                prompt_from_task=True,  # Set to True for prompt by task_name
+            ),
+        ),
+        weight_loader=weight_loaders.CheckpointWeightLoader("gs://openpi-assets/checkpoints/pi05_base/params"),
+        seed=1,
+        batch_size=256,
+        fsdp_devices=2,
+        num_train_steps=60000,
+    ),
+    TrainConfig(
+        name="pi05_base_aloha_full_sim_arx-x5_seed_2",
+        model=pi0_config.Pi0Config(pi05=True),
+        data=LeRobotAlohaDataConfig(
+            repo_id="RoboDojo_sim_arx-x5_v30",
+            assets=AssetsConfig(
+                assets_dir=str(_ROBODOJO_ASSETS_DIR),
+                asset_id="arx_x5_sim",
+            ),
+            repack_transforms=_transforms.Group(
+                inputs=[
+                    _transforms.RepackTransform(
+                        {
+                            "images": {
+                                "cam_high": "observation.images.cam_high",
+                                "cam_left_wrist": "observation.images.cam_left_wrist",
+                                "cam_right_wrist": "observation.images.cam_right_wrist",
+                            },
+                            "state": "observation.state",
+                            "actions": "action",
+                            "prompt": "prompt",
+                        }
+                    )
+                ]
+            ),
+            base_config=DataConfig(
+                prompt_from_task=True,  # Set to True for prompt by task_name
+            ),
+        ),
+        weight_loader=weight_loaders.CheckpointWeightLoader("gs://openpi-assets/checkpoints/pi05_base/params"),
+        seed=2,
+        batch_size=256,
+        fsdp_devices=2,
+        num_train_steps=60000,
+    ),
+    # Tianji Marvin + dual Wuji Hand Gen1 (54D). Set data.repo_id to your LeRobot dataset
+    # before compute_norm_stats / train. Skip RoboDojo process_data.py; use wuji mcap→LeRobot.
+    TrainConfig(
+        name="pi05_wuji_marvin_54d",
+        model=pi0_config.Pi0Config(pi05=True, action_dim=54, action_horizon=50, max_token_len=256),
+        data=LeRobotWujiDataConfig(
+            # Replace with your LeRobot repo id or local dataset id under HF_LEROBOT_HOME.
+            repo_id="tianji_marvin_wuji",
+            assets=AssetsConfig(
+                assets_dir=str(_WUJI_ASSETS_DIR),
+                asset_id="tianji_marvin_wuji",
+            ),
+            base_config=DataConfig(
+                prompt_from_task=True,
+            ),
+            extra_delta_transform=True,
+            # If your dataset uses cam_high instead of stereo_right, set:
+            # base_image_key="observation.images.cam_high",
+        ),
+        weight_loader=weight_loaders.PartialCheckpointWeightLoader(
+            "gs://openpi-assets/checkpoints/pi05_base/params"
+        ),
+        num_train_steps=30_000,
+        batch_size=64,
+        lr_schedule=_optimizer.CosineDecaySchedule(
+            warmup_steps=1_000,
+            peak_lr=5e-5,
+            decay_steps=30_000,
+            decay_lr=5e-6,
+        ),
+    ),
+    # Pika dual-arm T-shirt folding: 14-D absolute next-state targets at 10 Hz.
+    TrainConfig(
+        name="pi05_pika_bimanual_fisheye",
+        model=pi0_config.Pi0Config(pi05=True, action_dim=14, action_horizon=16, max_token_len=256),
+        data=LeRobotPikaBimanualDataConfig(
+            repo_id="pika/fold_cloth_0910_bimanual_fisheye",
+            # Must match compute_norm_stats.py, which writes below repo_id.
+            assets=AssetsConfig(asset_id="pika/fold_cloth_0910_bimanual_fisheye"),
+        ),
+        weight_loader=weight_loaders.PartialCheckpointWeightLoader(
+            "gs://openpi-assets/checkpoints/pi05_base/params"
+        ),
+        # The launchers set this to pika_manip/data/openpi_assets. Keeping the
+        # default permits a standalone OpenPI checkout to use ./assets instead.
+        assets_base_dir=os.environ.get("PIKA_OPENPI_ASSETS_BASE_DIR", "./assets"),
+        num_train_steps=30_000,
+        batch_size=64,
+        num_workers=4,
+        lr_schedule=_optimizer.CosineDecaySchedule(
+            warmup_steps=1_000,
+            peak_lr=5e-5,
+            decay_steps=30_000,
+            decay_lr=5e-6,
+        ),
+    ),
+    # Single-GPU-friendly LoRA alternative to the full Pika fine-tune above.
+    # It reuses the same local 50-episode train split and normalization stats.
+    TrainConfig(
+        name="pi05_pika_bimanual_fisheye_lora",
+        model=_PIKA_BIMANUAL_LORA_MODEL,
+        freeze_filter=nnx.Any(
+            _PIKA_BIMANUAL_LORA_MODEL.get_freeze_filter(),
+            # The ViT is not a LoRA module in OpenPI. Freeze it explicitly so
+            # this is adapter + Pika-projection tuning rather than an implicit
+            # full visual-backbone fine-tune.
+            nnx_utils.PathRegex("PaliGemma/img/.*"),
+        ),
+        data=LeRobotPikaBimanualDataConfig(
+            repo_id="pika/fold_cloth_0910_bimanual_fisheye",
+            # Reuse stats generated by prepare_pi05_pika_bimanual.sh for the
+            # full configuration; data semantics are identical.
+            assets=AssetsConfig(
+                assets_dir=os.path.join(
+                    os.environ.get("PIKA_OPENPI_ASSETS_BASE_DIR", "./assets"),
+                    "pi05_pika_bimanual_fisheye",
+                ),
+                asset_id="pika/fold_cloth_0910_bimanual_fisheye",
+            ),
+        ),
+        weight_loader=weight_loaders.PartialCheckpointWeightLoader(
+            "gs://openpi-assets/checkpoints/pi05_base/params"
+        ),
+        # EMA would update and retain a second copy of every frozen base weight
+        # on every step. It is prohibitively expensive for single-GPU LoRA and
+        # provides little value for this small, task-specific dataset.
+        ema_decay=None,
+        # Conservative starting point for one RTX 4090. The LoRA launcher
+        # defaults to this value but users may override BATCH_SIZE.
+        num_train_steps=30_000,
+        batch_size=4,
+        num_workers=4,
+        lr_schedule=_optimizer.CosineDecaySchedule(
+            warmup_steps=1_000,
+            peak_lr=5e-5,
+            decay_steps=30_000,
+            decay_lr=5e-6,
+        ),
+    ),
+    # Relative-pose Pika dual-arm configuration. The underlying LeRobot data
+    # deliberately remains lossless absolute pose data; the data transform
+    # converts it to a current-frame SE(3) representation before computing
+    # normalization statistics and before every training/inference call.
+    TrainConfig(
+        name="pi05_pika_bimanual_fisheye_relative",
+        model=pi0_config.Pi0Config(pi05=True, action_dim=14, action_horizon=16, max_token_len=256),
+        data=LeRobotPikaBimanualDataConfig(
+            repo_id="pika/fold_cloth_0910_bimanual_fisheye",
+            assets=AssetsConfig(asset_id="pika/fold_cloth_0910_bimanual_fisheye"),
+            relative_pose=True,
+        ),
+        weight_loader=weight_loaders.PartialCheckpointWeightLoader(
+            "gs://openpi-assets/checkpoints/pi05_base/params"
+        ),
+        assets_base_dir=os.environ.get("PIKA_OPENPI_ASSETS_BASE_DIR", "./assets"),
+        num_train_steps=30_000,
+        batch_size=64,
+        num_workers=4,
+        lr_schedule=_optimizer.CosineDecaySchedule(
+            warmup_steps=1_000,
+            peak_lr=5e-5,
+            decay_steps=30_000,
+            decay_lr=5e-6,
+        ),
+    ),
+    TrainConfig(
+        name="pi05_pika_bimanual_fisheye_relative_lora",
+        model=_PIKA_BIMANUAL_LORA_MODEL,
+        freeze_filter=nnx.Any(
+            _PIKA_BIMANUAL_LORA_MODEL.get_freeze_filter(),
+            nnx_utils.PathRegex("PaliGemma/img/.*"),
+        ),
+        data=LeRobotPikaBimanualDataConfig(
+            repo_id="pika/fold_cloth_0910_bimanual_fisheye",
+            assets=AssetsConfig(
+                assets_dir=os.path.join(
+                    os.environ.get("PIKA_OPENPI_ASSETS_BASE_DIR", "./assets"),
+                    "pi05_pika_bimanual_fisheye_relative",
+                ),
+                asset_id="pika/fold_cloth_0910_bimanual_fisheye",
+            ),
+            relative_pose=True,
+        ),
+        weight_loader=weight_loaders.PartialCheckpointWeightLoader(
+            "gs://openpi-assets/checkpoints/pi05_base/params"
+        ),
+        ema_decay=None,
+        num_train_steps=30_000,
+        batch_size=4,
+        num_workers=4,
+        lr_schedule=_optimizer.CosineDecaySchedule(
+            warmup_steps=1_000,
+            peak_lr=5e-5,
+            decay_steps=30_000,
+            decay_lr=5e-6,
+        ),
+    ),
+]
+
+if len({config.name for config in _CONFIGS}) != len(_CONFIGS):
+    raise ValueError("Config names must be unique.")
+_CONFIGS_DICT = {config.name: config for config in _CONFIGS}
+
+
+def cli() -> TrainConfig:
+    return tyro.extras.overridable_config_cli({k: (k, v) for k, v in _CONFIGS_DICT.items()})
+
+
+def get_config(config_name: str) -> TrainConfig:
+    """Get a config by name."""
+    if config_name not in _CONFIGS_DICT:
+        closest = difflib.get_close_matches(config_name, _CONFIGS_DICT.keys(), n=1, cutoff=0.0)
+        closest_str = f" Did you mean '{closest[0]}'? " if closest else ""
+        raise ValueError(f"Config '{config_name}' not found.{closest_str}")
+
+    return _CONFIGS_DICT[config_name]

real_inference/source/XPolicyLab/policy/Pi_05/openpi/src/openpi/training/data_loader.py

@@ -0,0 +1,549 @@
+from collections.abc import Iterator, Sequence
+import logging
+import multiprocessing
+import os
+import typing
+from typing import Literal, Protocol, SupportsIndex, TypeVar
+
+import jax
+import jax.numpy as jnp
+import numpy as np
+import torch
+
+try:
+    import lerobot.datasets.lerobot_dataset as lerobot_dataset
+except ModuleNotFoundError as exc:
+    if exc.name != "lerobot.datasets":
+        raise
+    import lerobot.common.datasets.lerobot_dataset as lerobot_dataset
+
+import openpi.models.model as _model
+import openpi.training.config as _config
+from openpi.training.droid_rlds_dataset import DroidRldsDataset
+import openpi.transforms as _transforms
+
+T_co = TypeVar("T_co", covariant=True)
+
+
+class Dataset(Protocol[T_co]):
+    """Interface for a dataset with random access."""
+
+    def __getitem__(self, index: SupportsIndex) -> T_co:
+        raise NotImplementedError("Subclasses of Dataset should implement __getitem__.")
+
+    def __len__(self) -> int:
+        raise NotImplementedError("Subclasses of Dataset should implement __len__.")
+
+
+class IterableDataset(Protocol[T_co]):
+    """Interface for an iterable dataset."""
+
+    def __iter__(self) -> Iterator[T_co]:
+        raise NotImplementedError("Subclasses of IterableDataset should implement __iter__.")
+
+    def __len__(self) -> int:
+        raise NotImplementedError("Subclasses of Dataset should implement __len__.")
+
+
+class DataLoader(Protocol[T_co]):
+    """Interface for a data loader."""
+
+    def data_config(self) -> _config.DataConfig:
+        """Get the data config for this data loader."""
+        raise NotImplementedError("Subclasses of DataLoader should implement data_config.")
+
+    def __iter__(self) -> Iterator[T_co]:
+        raise NotImplementedError("Subclasses of DataLoader should implement __iter__.")
+
+
+class TransformedDataset(Dataset[T_co]):
+    def __init__(self, dataset: Dataset, transforms: Sequence[_transforms.DataTransformFn]):
+        self._dataset = dataset
+        self._transform = _transforms.compose(transforms)
+
+    def __getitem__(self, index: SupportsIndex) -> T_co:
+        return self._transform(self._dataset[index])
+
+    def __len__(self) -> int:
+        return len(self._dataset)
+
+
+class IterableTransformedDataset(IterableDataset[T_co]):
+    def __init__(
+        self,
+        dataset: IterableDataset,
+        transforms: Sequence[_transforms.DataTransformFn],
+        *,
+        is_batched: bool = False,
+    ):
+        self._dataset = dataset
+        self._transform = _transforms.compose(transforms)
+        self._is_batched = is_batched
+
+    def __iter__(self):
+        for sample in self._dataset:
+            if self._is_batched:
+                # Transforms are designed to be applied to individual samples. So we need to split the batch into
+                # individual samples and apply the transform to each sample individually.
+                batch_size = next(v.shape[0] for v in sample.values())
+
+                # Split batch into individual samples using tree_map
+                individual_samples = [jax.tree.map(lambda x: x[i], sample) for i in range(batch_size)]  # noqa: B023
+
+                # Transform each sample
+                transformed = [self._transform(s) for s in individual_samples]
+
+                # Recombine batch with tree_map
+                yield jax.tree.map(lambda *x: np.stack(x, axis=0), *transformed)
+            else:
+                yield self._transform(sample)
+
+    def __len__(self) -> int:
+        return len(self._dataset)
+
+
+class FakeDataset(Dataset):
+    def __init__(self, model_config: _model.BaseModelConfig, num_samples: int):
+        self._num_samples = num_samples
+        self._observation_spec, self._action_spec = model_config.inputs_spec()
+
+    def __getitem__(self, index: SupportsIndex) -> dict:
+        rng = jax.random.key(index.__index__())
+
+        def make_from_spec(spec: jax.ShapeDtypeStruct):
+            nonlocal rng
+            rng, data_rng = jax.random.split(rng)
+            # Remove the batch dimension.
+            shape = spec.shape[1:]
+            if spec.dtype == jnp.float32:
+                return jax.random.uniform(data_rng, shape=shape, minval=-1.0, maxval=1.0)
+            if spec.dtype == jnp.int32:
+                return jax.random.randint(data_rng, shape=shape, minval=0, maxval=2048)
+            return jnp.zeros(shape=shape, dtype=spec.dtype)
+
+        observation = jax.tree.map(make_from_spec, self._observation_spec)
+        action = jax.tree.map(make_from_spec, self._action_spec)
+
+        return {
+            **observation.to_dict(),
+            "actions": action,
+        }
+
+    def __len__(self) -> int:
+        return self._num_samples
+
+
+def create_torch_dataset(
+    data_config: _config.DataConfig, action_horizon: int, model_config: _model.BaseModelConfig
+) -> Dataset:
+    """Create a dataset for training."""
+    repo_id = data_config.repo_id
+    if repo_id is None:
+        raise ValueError("Repo ID is not set. Cannot create dataset.")
+    if repo_id == "fake":
+        return FakeDataset(model_config, num_samples=1024)
+
+    dataset_meta = lerobot_dataset.LeRobotDatasetMetadata(repo_id, root=data_config.lerobot_root)
+    dataset = lerobot_dataset.LeRobotDataset(
+        data_config.repo_id,
+        root=data_config.lerobot_root,
+        episodes=list(data_config.episode_indices) if data_config.episode_indices is not None else None,
+        delta_timestamps={
+            key: [t / dataset_meta.fps for t in range(action_horizon)] for key in data_config.action_sequence_keys
+        },
+        video_backend=data_config.video_backend,
+    )
+
+    if data_config.prompt_from_task:
+        dataset = TransformedDataset(dataset, [_transforms.PromptFromLeRobotTask(dataset_meta.tasks)])
+
+    return dataset
+
+
+def create_rlds_dataset(
+    data_config: _config.DataConfig,
+    action_horizon: int,
+    batch_size: int,
+    *,
+    shuffle: bool = False,
+) -> Dataset:
+    # At the moment, we only support DROID for RLDS datasets.
+    return DroidRldsDataset(
+        data_dir=data_config.rlds_data_dir,
+        batch_size=batch_size,
+        shuffle=shuffle,
+        action_chunk_size=action_horizon,
+        action_space=data_config.action_space,
+        datasets=data_config.datasets,
+    )
+
+
+def transform_dataset(dataset: Dataset, data_config: _config.DataConfig, *, skip_norm_stats: bool = False) -> Dataset:
+    """Transform the dataset by applying the data transforms."""
+    norm_stats = {}
+    if data_config.repo_id != "fake" and not skip_norm_stats:
+        if data_config.norm_stats is None:
+            raise ValueError(
+                "Normalization stats not found. "
+                "Make sure to run `scripts/compute_norm_stats.py --config-name=<your-config>`."
+            )
+        norm_stats = data_config.norm_stats
+
+    return TransformedDataset(
+        dataset,
+        [
+            *data_config.repack_transforms.inputs,
+            *data_config.data_transforms.inputs,
+            _transforms.Normalize(norm_stats, use_quantiles=data_config.use_quantile_norm),
+            *data_config.model_transforms.inputs,
+        ],
+    )
+
+
+def transform_iterable_dataset(
+    dataset: IterableDataset,
+    data_config: _config.DataConfig,
+    *,
+    skip_norm_stats: bool = False,
+    is_batched: bool = False,
+) -> IterableDataset:
+    """Transform the dataset by applying the data transforms."""
+    norm_stats = {}
+    if data_config.repo_id != "fake" and not skip_norm_stats:
+        if data_config.norm_stats is None:
+            raise ValueError(
+                "Normalization stats not found. "
+                "Make sure to run `scripts/compute_norm_stats.py --config-name=<your-config>`."
+            )
+        norm_stats = data_config.norm_stats
+
+    return IterableTransformedDataset(
+        dataset,
+        [
+            *data_config.repack_transforms.inputs,
+            *data_config.data_transforms.inputs,
+            _transforms.Normalize(norm_stats, use_quantiles=data_config.use_quantile_norm),
+            *data_config.model_transforms.inputs,
+        ],
+        is_batched=is_batched,
+    )
+
+
+def create_data_loader(
+    config: _config.TrainConfig,
+    *,
+    sharding: jax.sharding.Sharding | None = None,
+    shuffle: bool = False,
+    num_batches: int | None = None,
+    skip_norm_stats: bool = False,
+    framework: Literal["jax", "pytorch"] = "jax",
+) -> DataLoader[tuple[_model.Observation, _model.Actions]]:
+    """Create a data loader for training.
+
+    Args:
+        config: The training configuration.
+        sharding: The sharding to use for the data loader (JAX only).
+        shuffle: Whether to shuffle the data.
+        num_batches: Determines the number of batches to return.
+        skip_norm_stats: Whether to skip data normalization.
+        framework: The framework to use ("jax" or "pytorch").
+    """
+    data_config = config.data.create(config.assets_dirs, config.model)
+    logging.info(f"data_config: {data_config}")
+
+    if data_config.rlds_data_dir is not None:
+        return create_rlds_data_loader(
+            data_config,
+            action_horizon=config.model.action_horizon,
+            batch_size=config.batch_size,
+            sharding=sharding,
+            shuffle=shuffle,
+            num_batches=num_batches,
+            skip_norm_stats=skip_norm_stats,
+            framework=framework,
+        )
+    return create_torch_data_loader(
+        data_config,
+        model_config=config.model,
+        action_horizon=config.model.action_horizon,
+        batch_size=config.batch_size,
+        sharding=sharding,
+        shuffle=shuffle,
+        num_batches=num_batches,
+        num_workers=config.num_workers,
+        seed=config.seed,
+        skip_norm_stats=skip_norm_stats,
+        framework=framework,
+    )
+
+
+def create_torch_data_loader(
+    data_config: _config.DataConfig,
+    model_config: _model.BaseModelConfig,
+    action_horizon: int,
+    batch_size: int,
+    *,
+    sharding: jax.sharding.Sharding | None = None,
+    skip_norm_stats: bool = False,
+    shuffle: bool = False,
+    num_batches: int | None = None,
+    num_workers: int = 0,
+    seed: int = 0,
+    framework: str = "jax",
+) -> DataLoader[tuple[_model.Observation, _model.Actions]]:
+    """Create a data loader for training.
+
+    Args:
+        data_config: The data configuration.
+        action_horizon: The action horizon.
+        batch_size: The batch size.
+        sharding: The sharding to use for the data loader. If None, the data loader will
+            use a single device sharding.
+        skip_norm_stats: Whether to skip data normalization.
+        shuffle: Whether to shuffle the data.
+        num_batches: Determines the number of batches to return. If the number exceeds the
+            number of batches in the dataset, the data loader will loop over the dataset.
+            If not provided, will iterate over the dataset indefinitely.
+        num_workers: The number of worker processes to use. If zero, the data loader will
+            execute in the main process.
+        seed: The seed to use for shuffling the data.
+    """
+    dataset = create_torch_dataset(data_config, action_horizon, model_config)
+    dataset = transform_dataset(dataset, data_config, skip_norm_stats=skip_norm_stats)
+
+    # Use TorchDataLoader for both frameworks
+    # For PyTorch DDP, create DistributedSampler and divide batch size by world size
+    # For JAX, divide by process count
+    sampler = None
+    if framework == "pytorch":
+        if torch.distributed.is_initialized():
+            sampler = torch.utils.data.distributed.DistributedSampler(
+                dataset,
+                num_replicas=torch.distributed.get_world_size(),
+                rank=torch.distributed.get_rank(),
+                shuffle=shuffle,
+                drop_last=True,
+            )
+            local_batch_size = batch_size // torch.distributed.get_world_size()
+        else:
+            local_batch_size = batch_size
+    else:
+        local_batch_size = batch_size // jax.process_count()
+
+    logging.info(f"local_batch_size: {local_batch_size}")
+    data_loader = TorchDataLoader(
+        dataset,
+        local_batch_size=local_batch_size,
+        sharding=None if framework == "pytorch" else sharding,
+        shuffle=(sampler is None and shuffle),  # Don't shuffle if using sampler
+        sampler=sampler,
+        num_batches=num_batches,
+        num_workers=num_workers,
+        seed=seed,
+        framework=framework,
+    )
+
+    return DataLoaderImpl(data_config, data_loader)
+
+
+def create_rlds_data_loader(
+    data_config: _config.DataConfig,
+    action_horizon: int,
+    batch_size: int,
+    *,
+    sharding: jax.sharding.Sharding | None = None,
+    skip_norm_stats: bool = False,
+    shuffle: bool = False,
+    num_batches: int | None = None,
+    framework: str = "jax",
+) -> DataLoader[tuple[_model.Observation, _model.Actions]]:
+    """Create an RLDS data loader for training.
+
+    Note: This data loader requires some extra dependencies -- see examples/droid/README_train.md
+
+    Args:
+        data_config: The data configuration.
+        action_horizon: The action horizon.
+        batch_size: The batch size.
+        sharding: The sharding to use for the data loader. If None, the data loader will
+            use a single device sharding.
+        skip_norm_stats: Whether to skip data normalization.
+        shuffle: Whether to shuffle the data.
+        num_batches: Determines the number of batches to return. If the number exceeds the
+            number of batches in the dataset, the data loader will loop over the dataset.
+            If not provided, will iterate over the dataset indefinitely.
+    """
+    if framework == "pytorch":
+        raise NotImplementedError("PyTorch RLDS data loader is not supported yet")
+    dataset = create_rlds_dataset(data_config, action_horizon, batch_size, shuffle=shuffle)
+    dataset = transform_iterable_dataset(dataset, data_config, skip_norm_stats=skip_norm_stats, is_batched=True)
+
+    data_loader = RLDSDataLoader(
+        dataset,
+        sharding=sharding,
+        num_batches=num_batches,
+    )
+
+    return DataLoaderImpl(data_config, data_loader)
+
+
+class TorchDataLoader:
+    """Torch data loader implementation."""
+
+    def __init__(
+        self,
+        dataset,
+        local_batch_size: int,
+        *,
+        sharding: jax.sharding.Sharding | None = None,
+        shuffle: bool = False,
+        sampler: torch.utils.data.Sampler | None = None,
+        num_batches: int | None = None,
+        num_workers: int = 0,
+        seed: int = 0,
+        framework: str = "jax",
+    ):
+        """Create a PyTorch data loader.
+
+        Args:
+            dataset: The dataset to load.
+            local_batch_size: The local batch size for each process.
+            sharding: The sharding to use for the data loader.
+            shuffle: Whether to shuffle the data.
+            num_batches: If provided, determines the number of returned batches. If the
+                number is larger than the number of batches in the dataset, the data loader
+                will loop over the dataset. If not provided, will iterate over the dataset
+                indefinitely.
+            num_workers: The number of worker processes to use. If zero, the data loader will
+                execute in the main process.
+            seed: The seed to use for shuffling the data.
+        """
+        if jax.process_count() > 1:
+            raise NotImplementedError("Data loading with multiple processes is not supported.")
+
+        if len(dataset) < local_batch_size:
+            raise ValueError(f"Local batch size ({local_batch_size}) is larger than the dataset size ({len(dataset)}).")
+
+        # Store sharding - None for PyTorch, JAX sharding for JAX
+        self._sharding = sharding
+        if sharding is None and framework == "jax":
+            # Use data parallel sharding by default for JAX only.
+            self._sharding = jax.sharding.NamedSharding(
+                jax.sharding.Mesh(jax.devices(), ("B",)),
+                jax.sharding.PartitionSpec("B"),
+            )
+        self._num_batches = num_batches
+
+        mp_context = None
+        if num_workers > 0:
+            mp_context = multiprocessing.get_context("spawn")
+
+        generator = torch.Generator()
+        generator.manual_seed(seed)
+        self._data_loader = torch.utils.data.DataLoader(
+            typing.cast(torch.utils.data.Dataset, dataset),
+            batch_size=local_batch_size,
+            shuffle=(sampler is None and shuffle),  # Don't shuffle if using sampler
+            sampler=sampler,
+            num_workers=num_workers,
+            multiprocessing_context=mp_context,
+            persistent_workers=num_workers > 0,
+            collate_fn=_collate_fn,
+            worker_init_fn=_worker_init_fn,
+            drop_last=True,
+            generator=generator,
+        )
+
+    @property
+    def torch_loader(self) -> torch.utils.data.DataLoader:
+        return self._data_loader
+
+    def __iter__(self):
+        num_items = 0
+        while True:
+            data_iter = iter(self._data_loader)
+            while True:
+                if self._num_batches is not None and num_items >= self._num_batches:
+                    return
+                try:
+                    batch = next(data_iter)
+                except StopIteration:
+                    break  # We've exhausted the dataset. Create a new iterator and start over.
+                num_items += 1
+                # For JAX, convert to sharded arrays; for PyTorch, return torch tensors
+                if self._sharding is not None:
+                    yield jax.tree.map(lambda x: jax.make_array_from_process_local_data(self._sharding, x), batch)
+                else:
+                    yield jax.tree.map(torch.as_tensor, batch)
+
+
+def _collate_fn(items):
+    """Collate the batch elements into batched numpy arrays."""
+    # Make sure to convert to numpy arrays before stacking since some of the incoming elements
+    # may be JAX arrays.
+    return jax.tree.map(lambda *xs: np.stack([np.asarray(x) for x in xs], axis=0), *items)
+
+
+def _worker_init_fn(worker_id: int) -> None:
+    """Tell JAX inside the worker process not to preallocate the GPU memory."""
+    # NOTE: This is called after jax is imported inside the worker process. This
+    # means that this approach will not work for selecting the backend.
+    os.environ["XLA_PYTHON_CLIENT_PREALLOCATE"] = "false"
+    os.environ["XLA_PYTHON_CLIENT_ALLOCATOR"] = "platform"
+
+
+class RLDSDataLoader:
+    """Shallow wrapper around the DROID data loader to make it compatible with openpi.
+
+    All batching already happens in the DROID dataset, so we don't need to do anything here.
+    """
+
+    def __init__(
+        self,
+        dataset: DroidRldsDataset,
+        *,
+        sharding: jax.sharding.Sharding | None = None,
+        num_batches: int | None = None,
+    ):
+        self._dataset = dataset
+        self._num_batches = num_batches
+
+        if jax.process_count() > 1:
+            raise NotImplementedError("Data loading with multiple processes is not supported.")
+
+        if sharding is None:
+            # Use data parallel sharding by default.
+            sharding = jax.sharding.NamedSharding(
+                jax.sharding.Mesh(jax.devices(), ("B",)),
+                jax.sharding.PartitionSpec("B"),
+            )
+
+        self._sharding = sharding
+        self._num_batches = num_batches
+
+    def __iter__(self):
+        num_items = 0
+        while True:
+            data_iter = iter(self._dataset)
+            while True:
+                if self._num_batches is not None and num_items >= self._num_batches:
+                    return
+                try:
+                    batch = next(data_iter)
+                except StopIteration:
+                    break  # We've exhausted the dataset. Create a new iterator and start over.
+                num_items += 1
+                yield jax.tree.map(lambda x: jax.make_array_from_process_local_data(self._sharding, x), batch)
+
+
+class DataLoaderImpl(DataLoader):
+    def __init__(self, data_config: _config.DataConfig, data_loader: TorchDataLoader | RLDSDataLoader):
+        self._data_config = data_config
+        self._data_loader = data_loader
+
+    def data_config(self) -> _config.DataConfig:
+        return self._data_config
+
+    def __iter__(self):
+        for batch in self._data_loader:
+            yield _model.Observation.from_dict(batch), batch["actions"]

real_inference/source/XPolicyLab/policy/Pi_05/openpi/src/openpi/training/data_loader_test.py

@@ -0,0 +1,84 @@
+import dataclasses
+
+import jax
+
+from openpi.models import pi0_config
+from openpi.training import config as _config
+from openpi.training import data_loader as _data_loader
+
+
+def test_torch_data_loader():
+    config = pi0_config.Pi0Config(action_dim=24, action_horizon=50, max_token_len=48)
+    dataset = _data_loader.FakeDataset(config, 16)
+
+    loader = _data_loader.TorchDataLoader(
+        dataset,
+        local_batch_size=4,
+        num_batches=2,
+    )
+    batches = list(loader)
+
+    assert len(batches) == 2
+    for batch in batches:
+        assert all(x.shape[0] == 4 for x in jax.tree.leaves(batch))
+
+
+def test_torch_data_loader_infinite():
+    config = pi0_config.Pi0Config(action_dim=24, action_horizon=50, max_token_len=48)
+    dataset = _data_loader.FakeDataset(config, 4)
+
+    loader = _data_loader.TorchDataLoader(dataset, local_batch_size=4)
+    data_iter = iter(loader)
+
+    for _ in range(10):
+        _ = next(data_iter)
+
+
+def test_torch_data_loader_parallel():
+    config = pi0_config.Pi0Config(action_dim=24, action_horizon=50, max_token_len=48)
+    dataset = _data_loader.FakeDataset(config, 10)
+
+    loader = _data_loader.TorchDataLoader(dataset, local_batch_size=4, num_batches=2, num_workers=2)
+    batches = list(loader)
+
+    assert len(batches) == 2
+
+    for batch in batches:
+        assert all(x.shape[0] == 4 for x in jax.tree.leaves(batch))
+
+
+def test_with_fake_dataset():
+    config = _config.get_config("debug")
+
+    loader = _data_loader.create_data_loader(config, skip_norm_stats=True, num_batches=2)
+    batches = list(loader)
+
+    assert len(batches) == 2
+
+    for batch in batches:
+        assert all(x.shape[0] == config.batch_size for x in jax.tree.leaves(batch))
+
+    for _, actions in batches:
+        assert actions.shape == (config.batch_size, config.model.action_horizon, config.model.action_dim)
+
+
+def test_with_real_dataset():
+    config = _config.get_config("pi0_aloha_sim")
+    config = dataclasses.replace(config, batch_size=4)
+
+    loader = _data_loader.create_data_loader(
+        config,
+        # Skip since we may not have the data available.
+        skip_norm_stats=True,
+        num_batches=2,
+        shuffle=True,
+    )
+    # Make sure that we can get the data config.
+    assert loader.data_config().repo_id == config.data.repo_id
+
+    batches = list(loader)
+
+    assert len(batches) == 2
+
+    for _, actions in batches:
+        assert actions.shape == (config.batch_size, config.model.action_horizon, config.model.action_dim)